Driving behavior-adaptive particle emissions in plug-in hybrid electric vehicles: cumulative-transient characteristics and clustered patterns for real- world monitoring

Driving behavior-adaptive particle emissions in plug-in hybrid electric vehicles: cumulative-transient characteristics and clustered patterns for real- world monitoring | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Driving behavior-adaptive particle emissions in plug-in hybrid electric vehicles: cumulative-transient characteristics and clustered patterns for real- world monitoring Ruizhi Huang, Yuzhuang Pian, Li Li, Yonghong Liu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7298392/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 03 Mar, 2026 Read the published version in Environmental Monitoring and Assessment → Version 1 posted 10 You are reading this latest preprint version Abstract Regulatory gaps in restart and cold/hot start emissions overlooked by current periodic technical inspection (PTI), and driving behaviors significantly impact plug-in hybrid electric vehicle (PHEV) particle number (PN) emissions under real driving conditions. Using portable emissions measurement systems (PEMS), this study building cumulative PN emissions across key segments (cold-start, restart) and instantaneous high-emission events across four distinct behaviors. Key findings reveal that calm and normal driving elevate cold-start PN (up to 6.2×10¹¹ #/km) due to prolonged engine-off intervals and slow warm-up. Aggressive driving’s frequent restarts yield lower per-event emissions owing to thermal advantages. Adaptive cruise control (ACC) minimizes total PN by combining thermally efficient engine operation with extended zero-emission phases (16–17% duration). Crucially, instantaneous high-emission analysis shows > 80% of PN concentrates in < 10% of driving duration, with emission thresholds varying dramatically (82-1366%) across behaviors—primarily due to divergent dominant modes favored by each behavior. To quantify these behavior-specific modes and their parametric signatures, k-means clustering was applied, and found distinct behavioral associations: aggressive driving predominantly linked to high-load/high-rpm operation (> 2800 rpm or > 80% load), while calm/normal driving elevates cold-start and restart contributions. Consequently, real-world emission monitoring necessitates behavior-adaptive dynamic scenarios, tailoring test focus and parametric design informed by clustered thresholds. plug-in hybrid electric vehicle (PHEV) particle number (PN) emission driving behavior real driving emission (RDE) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction With persistent environmental challenges and energy crises, the development of new energy vehicles has emerged as a pivotal strategy (Du et al., 2017 ). PHEVs mitigate range anxiety via dual-power systems compared to internal combustion engine vehicles (ICEVs) (Prati et al., 2021 ; F. Zhu et al., 2024 ), yet suffer frequent engine restarts that elevate PN emissions: under state of charge (SoC) sustain state in RDE, the number of restart for gasoline direct injection (GDI) PHEV turns to 140–190/100 km and the engine-on total time is around 0.5, and the emission factor ranges from 3.0-3.7 \(\:\text{×}{\text{10}}^{\text{11}}\:\) # /km, which is close to port fuel injection (PFI) ICEVs (Fan et al., 2020 ; Karjalainen et al., 2024 ; Yang et al., 2019 ). To address lifecycle emissions, the PTI, a European regulatory framework primarily targeting diesel verification, is being adapted for PHEVs in progressive regions. However, current monitoring primarily focuses on idle conditions, which fails to accurately reflect dynamic emission events like cold starts and restarts that dominate real-world PN profiles (Bainschab et al., 2020 ; Melas et al., 2023 ; Miao et al., 2024 ). Consequently, many studied focused on the PN characteristics of PHEV, which can instruct the design of dynamic scenarios for PTI, but most of them have identified SoC and battery ageing as significant factors to emission (Liu et al., 2024 ; Pavlovic et al., 2024 ), few studies have considered the impact of driving behaviors. For ICEVs, driving behaviors create varying power demands in the same location, thereby influencing emissions (Fernandes et al., 2021 ; Huang et al., 2021 ; Xie et al., 2024 ): Chandrashekar et al. found that aggressive driving were 18–30% higher than the normal driving for CO, HC, CO2, and NOx emission rates (Chandrashekar et al., 2023 ). For PHEVs, driving behaviors are still significant, influencing the demand power and battery state (Li et al., 2024 ). Few studies considered ACC behavior, which has shorter reacting times compared to human behaviors (Stogios et al., 2019 ), and may have different characteristic. Besides, most studies have focused the effect of behaviors on emission factors and road types, neglecting instantaneous characteristic and specific segment, such as cold start, hot start (China Ⅵ regulation defines hot start when initial coolant temperature exceeds 40℃) and restart, which have significant impacts: 70% of total PN occur within 30 s after restart in hybrid electric vehicles (HEVs), and the emission factor of cold start is 4.2 \(\:\text{×}{\text{10}}^{\text{11}}\:\) # /km, much higher than average value of total test (4.2 \(\:\text{×}{\text{10}}^{\text{10}}\:\) # /km) (Karjalainen et al., 2024 ; Park et al., 2025 ; Yao et al., 2025 ). Moreover, the particle size of particulate matter is another focus of emission characteristics. Most studies concentrate on PN which larger than 23 nm in size, and could not present PN factually: for HEVs, particles sized between 10–23 nm account for over 50% of total emissions (Wang et al., 2023 ). For GDI IECVs, particles sized between 6–23 nm account for 43–58% quantity of total emissions, and nucleation mode particles dominate emission of PFI vehicles (Hu et al., 2020 ; Miao et al., 2024 ; R. Zhu et al., 2024 ). Overall, current research overlooks two critical dimensions: behavior-specific mechanisms during high-emission transients, and emission characteristics of sub-23 nm particles that dominate nucleation-mode emissions. To address this, the study analyzes instantaneous and cumulative PN characteristics with smaller-sized particles across four typical driving behaviors, elucidating their impact mechanisms on real world PN emissions. Through cluster analysis, dominant high-emission patterns are classified and quantified, enabling the design of tailored dynamic monitoring scenarios, including next-generation PTI. The core contributions can be summarized as follows: The effect of driving behaviors on cumulative PN emissions, including total test, cold start and engine restart by calculating emission factor and analyzing relevant parameters. The effect of driving behaviors on instantaneous high emission by analyzing engine operations. Employing the k-means clustering to aggregates high-emission patterns. It identifies metrics characterizing dominant high emission patterns across diverse driving behaviors, quantifies parametric thresholds of typical behaviors and enable behavior-adaptive monitoring scenarios. To the best of our knowledge, this is the first work to evaluate the effect of driving behaviors on PN emission of a representative PHEV in real world conditions. Materials and methods This section first introduces the emission measurement system, test vehicle specifications, and experimental route design. Subsequently, we detail the calculation methodology for emission factors and analytical method for instantaneous PN rates. Emission measurement RDE of exhaust pollutants, including PN and CO, were measured by AVL M.O.V. E PEMS. This system comprised three core modules: gas analysis module (GAS PEMS AVL493), PN analysis module (PM PEMS AVL494) and exhaust flow meter (AVL495 EFM). As Fig. 1 shows, the PEMS system was installed at the back of the vehicle and an extension to the exhaust tailpipe was used to mount an exhaust flow meter (EFM). Exhaust gas temperature (EGT) sensor was installed in the extension of exhaust tailpipe, directly below the vehicle. Pretest and post-test calibration audits on PEMS were performed to verify the precision of PN emission. Other accessories included a GPS module and a OBD communication module, which could record the vehicle location information with 1 Hz sampling frequency, such as coolant temperature, engine speed and vehicle velocity. Test vehicle and route RDE tests were conducted in November 2023 in Chongqing, China. A commercially popular passenger car equipped with cooled exhaust gas recirculation (EGR) was selected for test. And the after-treatment device was a three-way catalyst (TWC) without a gasoline particulate filter (GPF). The vehicle was soaked under ambient conditions prior to testing, with auxiliary devices such as air conditioning turned off and total weight maintained consistently across all tests. Vehicle specifications were are detailed in Table S1 . Notably, due to high ambient temperatures and insufficient soaking time between consecutive tests, the initial coolant temperatures for calm, aggressive and ACC tests were higher. As Fig. 2 shows, the testing route almost covered the total urban area around the city. The testing duration of each test was between 6000–11000 s, with a fixed testing distance of 114.9 km. As Table S2 shows, the initial SoC was maintained at approximately 50% under charge-sustaining conditions. Although the SOC at the end of each test exhibits variability, fuel-equivalent conversion analysis reveals that the battery energy is significantly lower than fuel consumption, accounting for 0–11% of total energy expenditure, which is practically negligible. All tests were driven by different drivers, who were skilled and professional, and were instructed to keep their driving behaviors across the test. All driving behaviors were tested twice to ensure reproducibility. To qualify driving aggressiveness, some studies (Dhital et al., 2021 ; Prakash and Bodisco, 2019 ) used relative positive acceleration (RPA) to distinguish driving behaviors quantitatively. Fernandes (Fernandes et al., 2024 , 2023 ) used multiple vehicle operating variables to build a detecting and classifying model of driving behavior, and analyze the roundabouts by driving behaviors, operational performance and emissions. Thus, driving behaviors were divided based on RPA and va pos [95] in this research. The va pos [95] is the 95th percentile value of the second-by-second product of vehicle velocity and acceleration. As Fig. 2 shows, va pos [95] of aggressive driving behavior on rural and motorway slightly exceeded the defined threshold while RPA values of calm driving behavior in urban areas were marginally below the threshold. This indicated a clear distinction between different driving behaviors, covering a wide range of operational conditions. Notably, the differences between ACC and aggressive tests were more pronounced than those between normal and calm tests, highlighting significant variations in emission characteristics between aggressive and conservative driving behaviors. Emission factor and energy consumption calculation China Ⅵ regulation defines cold and hot start from engine’s first start, and it excludes cold start and vehicle stop status (velocity < 1 km/h) while calculating emission factor of total test. Thus, the emission factor \(\:{EF}_{i}\:\) (#/km) calculation can be divided into different phases: $$\:\begin{array}{c}{EF}_{i}\:\text{=}\frac{\sum\:_{\text{j=0}}^{\text{n}}{\text{q}}_{\text{j}}\text{×t}}{\text{3600×}\sum\:_{\text{j=0}}^{\text{n}}{\text{v}}_{\text{j}}\text{×t}}\#\left(1\right)\end{array}$$ In this formula, subscript \(\:i\) represents c, h, e for cold/hot start and total test respectively. \(\:{q}_{j}\:\) (#/s) is the instantaneous PN emission rate during section j and \(\:{v}_{j}\) (km/h) represents for driving velocity. By multiplying sampling interval \(\:t\) (1s) and summation, PN emission and total driving distance in each section can be calculated. The fuel-equivalent energy consumption \(\:E\) (kWh) is defined as: $$\:\begin{array}{c}\:E=\frac{V\times\:{q}_{g}\times\:{\eta\:}_{en}}{{\eta\:}_{em}}\#\left(2\right)\end{array}$$ where \(\:V\) represents the fuel consumption (L). \(\:{q}_{g\:}(L/kWh)\) , \(\:{n}_{en}\) , \(\:{n}_{em}\) denote the energy conversion factor, engine thermal efficiency and the transmission efficiency, with values of \(\:8.9\) , \(\:0.35\) and \(\:0.9\) respectively, which are derived from calorific value conversion and experimental data calculations. High emission sets (HES) on instantaneous PN rates High PN emission sets (HES), as defined by Mera (Mera et al., 2019 ), are used to characterize instantaneous emission: instantaneous PN rates are sorted in descending order, and cumulative emissions are calculated accordingly. With increase in the amount of computational data, the desired percentages of total emissions (20%, 50%, 80%) are reached, which are HES-20, HES-50 and HES-80. Because of the decreasing order of HES calculation, the minimum instantaneous PN emission rate in HES-20, HES-50, HES-80 are observed as emission threshold, which is corresponding to the intersection of curve and HES lines. As shown in Fig. S1 , HES-20, HES-50 and HES-80 represent the set of observations with the highest instantaneous PN emissions, which mean cumulative PN emissions are 20%, 50% and 80% of total. Besides, approximately 50% of the duration in each test had almost no emissions, and these periods are excluded from HES. For PFI HEV under RDE test, the PN thresholds of HES-20, HES-50, HES-80 are 27.36, 5.98, 1.14 \(\:\text{×}{\text{10}}^{\text{11}}\:\) # /s, and the time thresholds are 0.12, 0.56, 2.43%, revealing a extremely unbalance in time series (Wang et al., 2021 ). Results and discussion Cumulative PN characteristics in different driving behaviors In this section, we calculated segment-specific emissions and analyzed key parametric influences to examine cumulative PN characteristics under varying driving behaviors. As Table 1 and Fig. S2 show, emission factors throughout testing remained below China 6 regulatory limits ( \(\:6\text{×}{\text{10}}^{\text{11}}\:\) # /km), outperforming both GDI PHEVs under CS state and PFI HEVs (Kontses et al., 2020 ; Melas et al., 2022 ). However, emission varied substantially across driving behaviors: aggressive driving yielded the highest emissions, exceeding those of calm tests by a factor of two, while ACC achieved the lowest levels. Focusing on transient phases identified as critical PN sources in hybrid vehicles (Feinauer et al., 2022 ; Selleri et al., 2022 ), cumulative PN is categorized into restart (PN generated within the first 15 seconds of each engine restart event), cold/hot start and other operations. To resolve temporal overlaps (restarts occurring during cold/hot start), Table 2 distinguishes restart contribution excluding cold/hot start phases and total restart contribution including overlaps, which indicate isolated restart impact and comprehensive engine-start burden. The results indicate that driving behaviors significantly influence the PN portion of restart and cold start, with restart (except hot/cold start) emissions comprising 19.9–56.6% and cold start emissions ranging from 5.6–49.9% of the total test. As critical emission sources and behavior-dependent variables, these segments necessitate phase-specific analysis to decode behavioral impacts. Table 1 Total emissions, PN portion and parameters of different running segments. Driving behavior and test set Emission factor of total test ( \(\:\text{×}{\text{10}}^{\text{10}}\:\) #/km) Initial engine coolant temperature (℃) PN portion of cold/hot start (%) Restart contribution except hot/cold start phase (%) Total restart contribution including overlaps (%) Calm test 1 1.91 27 49.9 29.6 76.3 Calm test 2 2.16 45 3.1 33.0 35.8 Normal test 1 3.06 26 44.4 25.5 67.9 Normal test 2 3.91 21 41.8 28.1 59.2 ACC test 1 1.45 21 16.6 36.7 39.8 ACC test 2 1.44 45 8.0 56.6 63.3 Aggressive test 1 3.65 26 6.6 19.9 20.7 Aggressive test 2 4.12 43 5.6 26.6 32.0 Driving behavior and test set Number of restarts Total test duration (s) Time portion of pure electric driving (%) Emission factor of cold/hot start (#/km) Coolant temperature rise rate until 70°C (°C/s) Calm test 1 75 10024 53.2 5.0 \(\:\text{×}{\text{10}}^{\text{11}}\) 0.16 Calm test 2 77 8499 52.8 1.1 \(\:\text{×}{\text{10}}^{\text{11}}\) 0.17 Normal test 1 106 10628 46.5 6.2 \(\:\text{×}{\text{10}}^{\text{11}}\) 0.16 Normal test 2 100 8570 21.6 3.3 \(\:\text{×}{\text{10}}^{\text{11}}\) 0.14 ACC test 1 108 10977 29.8 8.0 \(\:\text{×}{\text{10}}^{\text{1}\text{0}}\) 0.29 ACC test 2 110 7926 44.2 4.6 \(\:\text{×}{\text{10}}^{\text{1}\text{0}}\) 0.23 Aggressive test 1 123 9600 51.6 7.7 \(\:\text{×}{\text{10}}^{\text{1}\text{0}}\) 0.27 Aggressive test 2 111 6684 48.1 1.2 \(\:\text{×}{\text{10}}^{\text{11}}\) 0.21 The impact of restart on PN emission Restart phases contribute dominantly (19.9–56.6%) and behavior-sensitively to PN. Previous studies (Park et al., 2025 ) indicate that HEV restart strategies optimize energy efficiency and battery life by reducing unnecessary engine operation under low-load conditions (Choi et al., 2021 ). Furthermore, restarts critically influence emissions through two thermal parameters: EGT, which determines catalyst conversion efficiency; engine-off interval, which governs catalyst cooling rate. Therefore, this section investigates the linkage between EGT and engine-off interval to examine how driving behaviors influence both restart strategy execution and resultant emissions. As Fig. 3 and Table 1 show, behavioral analysis reveals stark thermal divergences: aggressive driving shortened engine-off intervals by 50% versus calm driving while elevating restart EGT by 38–64%, resulting in 68–79% of restart PN concentrated within the initial 15s. ACC driving protocols extended engine-off intervals 39–67% beyond normal, yielding comparable EGT but 21–52°C lower per-restart emissions than calm driving. These thermal divergences fundamentally restructured PN profiles: aggressive driving elevated restart frequencies (111–123 times during 6684–9600 s of test duration) while it reduces per-event emission intensity: although frequent restarts raise cumulative PN counts, they depress average EGT per restart, thereby diminishing emission peaks. Crucially, under controlled SOC, high frequency restarts elevate temperature, enhancing catalyst conversion efficiency: the resultant emission reduction outweighs the particle increase from restart events. Consequently, aggression inversely correlates with restart-induced emissions. This redefines optimization priorities: extended engine-off intervals (> 5 min) causing critical catalyst cooling emerge as greater threats than restart frequency, demanding mitigation in next-generation energy management strategies. Impacts of engine cold/hot start Cold start phase contributes over 40% emission of total trips (Badshah et al., 2016 ; Park et al., 2025 ), yet critically overlooks the significant influence of driving behaviors (3.1–49.9%). In this section, this phase is dissected through two lenses: cold/hot start emission factor and contrasts and fixed engine warm-up strategy, including behavioral interference with thermal conditioning. As shown in Table 1 , cold start emission factors (0.77–6.2 \(\:\text{×}{\text{10}}^{\text{11}}\) #/km) consistently exceeded hot start levels (0.46–1.1 \(\:\text{×}{\text{10}}^{\text{11}}\) #/km). This disparity stems predominantly from sub-40°C catalyst inefficiency during cold start, where low EGT delay catalytic activation (Badshah et al., 2016 ). In contrast, hot start operations (initial coolant > 40°C) maintain high catalyst efficiency, minimizing behavior-dependent emission variations. As Fig. 4 shows, vehicle tends to maintain a low engine load (< 50%) during the initial engine start in each test, which is also found in HEVs and ICEVs (Chambon et al., 2013 ; Takagi et al., 2010 ). Low engine load falls within the optimal range of exhaust gas recirculation (EGR) rates. It would reduce the burning temperature and promote more uniform combustion, which inhibits the formation of large carbonaceous agglomerates and decreases nucleation mode emissions (Guo et al., 2023 ). Thus, this strategy helps maintain low emission rates while increasing TWC temperature, thereby reducing PN emission in subsequent engine restart. However, prioritizing fuel economy over emission reduction may compromise outcomes: low engine load reduced fuel atomization efficiency, and leads to incomplete combustion. Experimentally, this strategy was rigidly implemented, lasting precisely 60 seconds exclusively during the initial engine start of each test, irrespective of driving behavior, ambient conditions, or catalyst temperature. This phenomenon’s restriction to first-start events likely stems from insufficient engine off duration to trigger the strategy during subsequent cycles. Furthermore, calm and normal driving exhibit gradual power transitions and lower peak demands, slowing engine coolant temperature rise rate (0.14–0.17°C/s) and number of restart and elevating PN emissions. As noted in prior research (Hu et al., 2024 ), emissions inherently represent a trade-off relation with fuel economy under energy management strategies. Nevertheless, strategy designers may have underestimated the substantial impact of driving behavior on particle emissions during cold start phases, thereby contributing to the occurrence of high emission events. Instantaneous PN threshold and time proportion of HES As Fig. S3 and Table 2 show, the majority of emissions are generated within a small fraction of time. For normal and ACC tests, HES-20 are generated in less than 0.2% of test duration, and HES-80 are generated in less than 10% of test duration. Aggressive tests have a slightly higher time percentage, reaching 0.5% of duration for HES-20 and 20% for HES-80. Table 2 Thresholds of instantaneous PN rates and time proportion of HES in all tests. Driving behavior and test set Calm test 1 Normal test 1 ACC test 1 Aggressive test 1 Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) HES-20 84.4 0.1 98.5 0.1 9.4 0.2 10.2 0.4 HES-50 15.5 0.2 16.7 0.3 2.1 1.4 1.5 4.4 HES-80 5.6 3.8 0.9 3.7 0.4 7.3 0.4 20.4 Driving behavior and test set Calm test 2 Normal test 2 ACC test 2 Aggressive test 2 Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) Threshold ( \(\:{\text{10}}^{\text{9}}\text{}\text{\#/s}\) ) Time (%) Threshold ( \(\:{\text{10}}^{\text{9}}\text{\:\#/s}\) ) Time (%) HES-20 85.5 0.1 41.4 0.1 24.0 0.1 30.9 0.2 HES-50 9.4 0.5 7.5 1.1 4.2 0.7 3.7 2.9 HES-80 6.9 4.6 0.9 9.5 0.4 6.9 0.8 17.3 Furthermore, it is observed that for human driving behaviors (normal, aggressive, calm), the time proportion of HES exhibits a clear negative correlation with the PN rate threshold of HES. In normal test the PN rate threshold is 12.5-1013.3% higher than aggressive ones, and the time proportion is 82.1-1366.7% lower. Calm tests show similar trends to normal tests. With similar total emissions, Table 2 indicates aggressive tests do not have skyscraping emission rates compared with normal tests, and need more time to accumulate the same order of magnitude of emissions. The low cumulative emissions in ACC tests can be attributed to two factors: first, the PN thresholds for HES-20 and HES-80 are generally lower than those in aggressive tests while HES-50 thresholds are generally higher, suggesting a more uniform distribution of instantaneous PN rates. Besides, as shown in Fig. S3, when cumulative emissions of ACC tests reach 99.9%, the corresponding time percentages account for 84.0% and 82.5%. This indicates a significant proportion of time with almost no PN emissions, which is not entirely due to engine shutdown or electric driving. In summary, low PN emissions in ACC tests are a result of both lower peak rates and a higher time proportion of negligible emissions. Engine speed and load impact on driving behaviors Building upon earlier findings revealing consistent transient high-emission patterns across repeated tests under identical driving behaviors, this section consolidates tests with the same driving behavior to address the statistical limitations of HES-20 sampling. Engine operating status is considered the most relevant parameter influencing PN. Figure 5 illustrates the engine operation, mean values, frequency distribution of HES, and mean engine operation values across the total test (except vehicle stop and electric driving state). Holistically, despite identical routes and engine strategies, compared to normal and calm driving, aggressive and ACC driving exhibit higher power demands, resulting in elevated engine speed and load distributions (Δ = 80–179 rpm, 0–6% in mean value). Higher engine speed reduces fuel evaporation time, inducing localized incomplete combustion and increased PN emissions (Chen et al., 2017 ). Concurrently, high engine load accelerates combustion deterioration, amplifying full-spectrum particle emissions, particularly for particles under 23 nm (Hu et al., 2023 ). For HES distributions, all driving behaviors show concentrated occurrence at around 2000 rpm. However, aggressive and conservative behaviors exhibit greater divergence in mean HES engine speeds (Δ = 95-1094 rpm, 1–33%), attributable to secondary emission peaks near 3000 rpm in aggressive and ACC driving. Thus, these behaviors concentrate high-emission zones under high engine operation (> 50% of PN spikes at > 2500 rpm and > 80% load), corroborating prior conclusions. Besides, driving behavior emission differentials further correlate significantly with HES tendency of engine load and speed distribution. For calm and normal driving, HES-20 to HES-80 shows rising mean speed and load (194–394 rpm, 9–19%), indicating dominant emissions at lower speeds with secondary peaks at higher loads. This confirms cold start and restart emissions exceed those under stabilized high load conditions. For aggressive driving, HES-20 to HES-80 shows declining speed and load (252 rpm, 4%), demonstrating higher emissions during high load operation than others. For ACC driving, HES-20 to HES-80 exhibits a declining engine speed trajectory (54 rpm) coupled with rising load (9%), further evidencing higher emission peaks in high-speed regimes. Current PTI for emission control prioritize idle conditions, which contributes < 1% emission, misaligned with core high-emission engine states identified under RDE condition. This misalignment arises from PTI's static and low load testing versus real world dynamic high emission scenarios. Consequently, following the previous research of HES and high emission segments across diverse driving behaviors, these events are grouped by parameters to identify statistically robust high emission modes with defined parametric signatures. This enables quantitative linkage between emission patterns and driving behaviors, thereby facilitating behavior-differentiated dynamic inspection scenario design for emission-oriented vehicle monitoring, such as PTI. Clustering and behavior-adaptive monitoring for PN emissions Cluster analysis is recognized as an effective data processing method, widely adopted in emission characterization studies (Mehnatkesh et al., 2025 ). This study applies k-means clustering to HES. Temporally continuous high-emission points are statistically characterized and consolidated as input data, with rotational engine speed, load, EGT, and acceleration serving as clustering features. The optimal cluster count is adaptively determined by maximizing the silhouette coefficient. Consistent with the earlier methodology addressing insufficient transient emission samples, tests under the same behaviors are aggregated. We statistically characterized the engine speed, load, velocity and acceleration at cluster centroids, along with the proportions of cold start and restart events within each cluster. As Fig. 6 shows, when multiple clusters share the same emission mode, HES level and driving behavior, their parameter values (90th percentile) are averaged for visualization, and the cluster count per group is annotated above each radar plot. The emission modes are divided into three modes, the threshold of restart and cold start is > 80% event; the threshold of high engine operation is engine speed > 2800 rpm or load > 80%. Figure 6 and Table S3 reveal behavioral emission patterns consistent with prior findings: 77.8% of clusters are classified as high engine operation in aggressive driving, thus it predominantly concentrates high emission clusters in high engine operation segments. Within high engine operating clusters, speed and load distributions converge across behaviors, yet aggressive driving exhibits significantly higher acceleration (0.5–1.9 vs 0.0-0.7 m/s²). This substantiates its pronounced impact on transient power output, necessitating dynamic tests with steeper power transitions and higher acceleration. ACC driving distributes clusters across all three modes, but 45% of clusters are high engine operation. The remaining clusters occur in HES-20 cold start, restart and HES-80 restart modes, explaining the inverted speed-load trajectory (Section 3.2.2). Compare to aggressive driving, calm and normal driving demonstrate inverse distribution patterns: all clusters are cold start dominant in HES-20, over 57% of clusters are restart-dominant in HES-50. For HES-80, clusters are split between high engine operation and restart modes. Besides, calm driving shows greater conservatism, retaining cold start clusters at HES-50. These results validate the efficacy of the classification thresholds and significant behavioral influences on high emission segments. Table 3 Driving behavior-adaptive focus on dominant emission modes and transient characteristics. Driving behavior Dominant emission mode Full process and transient high emission focus Calm/ Normal Restart (46.1% of HES events in calm, 50.9% of HES events in normal) Average start-up EGT − 40%, lower restart frequency and longer test duration vs. aggressive tests. Calm/ Normal Cold start (1.6% of HES events in calm, 2.2% of HES events in normal) Prohibit restarts in first 60s, coolant temperature rise rate < 0.2°C/s, pay attention to the sudden increase in engine speed load after 60 seconds or the second restart. ACC Restart (36.6% of HES events) Average start-up EGT − 30% compare to aggressive tests. ACC High engine operation (59.1% of HES events) Engine speed threshold > 2800 rpm, engine load threshold > 80%, medium acceleration. Aggressive High engine operation (58.5% of HES events) Engine speed threshold > 2800 rpm, engine load threshold > 80%, average acceleration + 50% vs. ACC tests. As Table 3 shows, based on the above conclusions and analysis of high emission modes and HES events, the high emission parameter characteristics and emission profiles of PHEVs under different driving behaviors are summarized. Driving behaviors systematically reconfigure the parametric signatures and modal dominance hierarchies of emission-critical states, necessitating three key adaptations in dynamic monitoring design: temporal allocation must prioritize high engine operation for aggressive driving (58.5% of HES events) versus restart phases for conservative driving (46.1/50.9%); parameter scaling should reflect behavior-specific extremes, expanding acceleration ranges from 0.5 to 1.9 m/s² to capture aggressive transients. Critically, while behavioral variations reprioritize monitoring focus across modes, no emission critical state loses relevance. Thus, dynamic scenarios must retain but reweight all modal components based on driving behavior. Conclusion A representative hybrid system-equipped PHEV (EGR-enabled, GPF-free) was analyzed for cumulative and instantaneous PN characteristics across various driving behaviors under real world conditions, with operational parameters examined as influencing factors. Cumulative PN characteristics were evaluated across three specific segments: cold/hot start and restart. Normal and calm driving have higher PN portion from restart (19.9–56.6%) and cold start (41.8–49.9%) than aggressive behaviors. This is attributed to lower and smoother power demands, resulting in longer engine-off interval (53.2–74.1 s average) and lower initial restart EGT (62.0-70.6°C average). Thus, it is proved that extending engine-off intervals causing critical catalyst cooling emerge as greater threats than restart frequency for PN emission. ACC driving achieve the lowest total emissions because of significantly extended zero-emission durations (16.0-17.5%). While cold start engine strategies balance fuel economy and emission control through low-load operation (< 50%) to promote uniform combustion, the impact of driving behavior on thermal conditioning is overlook: aggressive electric assist during calm and normal driving slows coolant warming (0.14–0.17°C/s) and decrease the number of restart, exacerbating PN emission variance by up to 8-fold (7.7 \(\:\text{×}{\text{10}}^{\text{10}}\) -6.2 \(\:\text{×}{\text{10}}^{\text{1}\text{1}}\:\) #/km) across tests. Overall, the emission reduction potential of driving behavior can be quantitatively inferred. That is to say, the transition from radical to conservative driving can reduce PN by 54% and ACC by further reducing emission duration by 17%. For instantaneous emission, HES analysis reveals that over 80% of PN emissions concentrate in less than 10% of duration, with emission thresholds exhibiting profound behavioral divergence: aggressive driving requires 82–1366% lower instantaneous PN rates to trigger HES events compared to normal driving, yet sustains high emission phases 20.4% longer (HES-80). This volatility originates from behavioral modulation of engine operating states: aggressive driving concentrated 77.8% of high-emission events at high load and speed regimes (> 2800 rpm or > 80% load) while calm and normal driving prioritized cold start and restart contributions. These parametric divergences necessitate shifting focus from emission levels to engine dynamics. Cluster analysis (k-means) resolves this by categorizing 85% of clusters into three high emission modes, while exposing behavioral nuances within modes: identical high-load clusters show higher acceleration (50%) in aggressive driving, and cold start clusters exhibit slower coolant warming in conservative driving. Therefore, for behavior-adaptive monitoring, aggressive driving demands high proportion of high engine operation segments with higher acceleration (0.5–1.9 m/s²), triggered at > 2800 rpm and 80% load thresholds to capture their extreme emissions. Conservative driving requires extended restart and cold start segments, monitoring the sudden increase in engine speed load after 60 seconds or the second restart; ACC driving necessitates hybrid scenarios balancing restart transients (EGT + 10% compensation compare to conservative ones) and sustained high engine operation. The shift in emission monitoring (such as PTI) from static to dynamic paradigms is evident, and it would be better to incorporate driving behavior variability into emission regulations and certification protocols to ensure real-world compliance. Behavioral variations reprioritize monitoring focus across modes, but no emission critical mode loses relevance. Thus, dynamic scenarios must retain but reweight all modal components by temporal allocation and parameter scaling based on driving behavior. Declarations Competing interests The authors declare no competing interests. Funding This work was supported by the Fundamental Research Key Program of Shenzhen (JCYJ20241202130016022) and National Natural Science Foundation of China (41975165). Author Contribution Ruizhi Huang: Methodology, Formal analysis, Writing-original draft, Visualization; Yuzhuang Pian: Investigation, Data curation, Writing-Review & editing; Li Li: Project administration, Conceptualization; Yonghong Liu: Supervision, Funding acquisition, Resources. References Badshah, H., Kittelson, D., Northrop, W., 2016. Particle Emissions from Light-Duty Vehicles during Cold-Cold Start. SAE Int. J. Engine s 9, 1775–1785. https://doi.org/10.4271/2016-01-0997 Bainschab, M., Schriefl, M.A., Bergmann, A., 2020. Particle number measurements within periodic technical inspections: A first quantitative assessment of the influence of size distributions and the fleet emission reduction. Atmospheric Environment: X 8, 100095. https://doi.org/10.1016/j.aeaoa.2020.100095 Chambon, P., Deter, D., Irick, D.K., Smith, D.E., 2013. PHEV Cold Start Emissions Management. SAE International Journal of Alternative Powertrains 2, 252–260. https://doi.org/ 10.4271/2013-01-0358 Chandrashekar, C., Rawat, R.S., Chatterjee, P., Pawar, D.S., 2023. Evaluating the real-world emissions of diesel passenger Car in Indian heterogeneous traffic. Environ Monit Assess 195, 1248. https://doi.org/10.1007/s10661-023-11658-z Chen, L., Liang, Z., Zhang, X., Shuai, S., 2017. Characterizing particulate matter emissions from GDI and PFI vehicles under transient and cold start conditions. Fuel 189, 131–140. https://doi.org/10.1016/j.fuel.2016.10.055 Choi, Y., Yi, H., Oh, Y., Park, S., 2021. Effects of engine restart strategy on particle number emissions from a hybrid electric vehicle equipped with a gasoline direct injection engine. Atmospheric Environment 253, 118359. https://doi.org/10.1016/j.atmosenv.2021.118359 Dhital, N.B., Wang, S.-X., Lee, C.-H., Su, J., Tsai, M.-Y., Jhou, Y.-J., Yang, H.-H., 2021. Effects of driving behavior on real-world emissions of particulate matter, gaseous pollutants and particle-bound PAHs for diesel trucks. Environmental Pollution 286, 117292. https://doi.org/10.1016/j.envpol.2021.117292 Du, J., Ouyang, M., Chen, J., 2017. Prospects for Chinese electric vehicle technologies in 2016–2020: Ambition and rationality. Energy 120, 584–596. https://doi.org/10.1016/j.energy.2016.11.114 Fan, Q., Wang, Y., Xiao, J., Wang, Z., Li, W., Jia, T., Zheng, B., Taylor, R.I., 2020. Effect of Oil Viscosity and Driving Mode on Oil Dilution and Transient Emissions Including Particle Number in Plug-In Hybrid Electric Vehicle. SAE Technical Paper Series . https://doi.org/ 10.4271/2020-01-0362 Feinauer, M., Ehrenberger, S., Epple, F., Schripp, T., Grein, T., 2022. Investigating Particulate and Nitrogen Oxides Emissions of a Plug-In Hybrid Electric Vehicle for a Real-World Driving Scenario. Applied Sciences 12, 1404. https://doi.org/10.3390/app12031404 Fernandes, P., Ferreira, E., Macedo, E., Coelho, M.C., 2024. Unraveling roundabout dynamics: Analysis of driving behavior, vehicle performance, and exhaust emissions. Transportation Research Part D: Transport and Environment 133, 104308. https://doi.org/10.1016/j.trd.2024.104308 Fernandes, P., Macedo, E., Tomás, R., Coelho, M.C., 2023. Hybrid electric vehicle data-driven insights on hot-stabilized exhaust emissions and driving volatility. International Journal of Sustainable Transportation 18, 84–102. https://doi.org/10.1080/15568318.2023.2219629 Fernandes, P., Tomás, R., Ferreira, E., Bahmankhah, B., Coelho, M.C., 2021. Driving aggressiveness in hybrid electric vehicles: Assessing the impact of driving volatility on emission rates. Applied Energy 284, 116250. https://doi.org/10.1016/j.apenergy.2020.116250 Guo, Z., Wang, Boyuan, Liu, S., Zhang, Z., Wang, Buyu, Chang, C.-T., Wang, P., He, X., Sun, X., Shuai, S., 2023. Experimental investigation on emission characteristics of high reactivity gasoline compression ignition with different EGR and injection pressure strategies. Fuel 332, 126176. https://doi.org/10.1016/j.fuel.2022.126176 Hu, R., Chen, X., Li, L., Kong, F., Liu, Y., 2024. Exhaust emissions and energy conversion of hybrid and conventional CNG buses. Transportation Research Part D: Transport and Environment 135, 104405. https://doi.org/10.1016/j.trd.2024.104405 Hu, Z., Lu, Z., Song, B., Quan, Y., 2020. Impact of test cycle on mass, number and particle size distribution of particulates emitted from gasoline direct injection vehicles. The Science of the total environment 143128. https://doi.org/ 10.1016/j.scitotenv.2020.143128 Hu, Z., Xu, Y., Wang, Z., Zhang, H., Tan, P., Lou, D., 2023. An experimental study on particle number, micromorphology and nanostructure characteristics of particulate matter from a China Ⅵ gasoline direct injection engine. Atmospheric Environment: X . https://doi.org/ 10.1016/j.aeaoa.2023.100211 Huang, R., Ni, J., Cheng, Z.X., Wang, Q., Shi, X., Yao, X., 2021. Assessing the effects of ethanol additive and driving behaviors on fuel economy, particle number, and gaseous emissions of a GDI vehicle under real driving conditions. Fuel 306, 121642. https://doi.org/10.1016/j.envadv.2023.100454 Karjalainen, P., Leinonen, V., Olin, M., Vesisenaho, K., Marjanen, P., Järvinen, A., Simonen, P., Markkula, L., Kuuluvainen, H., Keskinen, J., Mikkonen, S., 2024. Real-world emissions of nanoparticles, particulate mass and black carbon from a plug-in hybrid vehicle compared to conventional gasoline vehicles. Environmental Advances 15, 100454. https://doi.org/10.1016/j.envadv.2023.100454 Kontses, A., Triantafyllopoulos, G., Ntziachristos, L., Samaras, Z., 2020. Particle number (PN) emissions from gasoline, diesel, LPG, CNG and hybrid-electric light-duty vehicles under real-world driving conditions. Atmospheric Environment 222, 117126.https://doi.org/10.1016/j.atmosenv.2019.117126 Li, B., Chen, Z., Zhu, X., Zhang, Z., Peng, Z., Zhao, H., He, H., 2024. Assessment of eco-driving strategies on carbon emissions for hybrid vehicles through portable emissions measurement systems. Atmospheric Pollution Research 102365. https://doi.org/10.1016/j.apr.2024.102365 Liu, Y., Li, C., Boger, T., Feng, X., Li, W., Chen, X., 2024. RDE PN Emission Challenges for a China 6 PHEV. SAE Technical Paper Series. https://doi.org/ 10.4271/2024-01-2386 Mehnatkesh, H., Gordon, D., Koch, C.R., 2025. Dynamic emission analysis of a hydrogen/diesel dual-fuel engine using clustering method. International Journal of Hydrogen Energy 136, 371–382. https://doi.org/10.1016/j.ijhydene.2025.04.385 Melas, A., Selleri, T., Franzetti, J., Ferrarese, C., Suarez-Bertoa, R., Giechaskiel, B., 2022. On-Road and Laboratory Emissions from Three Gasoline Plug-In Hybrid Vehicles-Part 2: Solid Particle Number Emissions. Energies 15, 5266. https://doi.org/10.3390/en15145266 Melas, A., Vasilatou, K., Suarez-Bertoa, R., Giechaskiel, B., 2023. Laboratory measurements with solid particle number instruments designed for periodic technical inspection (PTI) of vehicles. Measurement 215, 112839. https://doi.org/10.1016/j.measurement.2023.112839 Mera, Z., Fonseca, N., López, J.-M., Casanova, J., 2019. Analysis of the high instantaneous NOx emissions from Euro 6 diesel passenger cars under real driving conditions. Applied Energy 242, 1074–1089. https://doi.org/10.1016/j.apenergy.2019.03.120 Miao, X., Zhang, X., Wang, C., Li, J., Zhao, J., Qu, L., Liu, Y., Qi, S., Li, H., Fu, M., Jin, T., 2024. Effects of fuel and driving conditions on particle number emissions of China-VI gasoline vehicles: based on corrections to test results. Environ Monit Assess 196, 591. https://doi.org/10.1007/s10661-024-12756-2 Park, J., Kim, H., Kim, Y., Park, S., 2025. Real-world particle number emissions from hybrid electric vehicles with port fuel injection and dual injection (MPI-GDI) systems. Journal of Cleaner Production 510, 145644. https://doi.org/10.1016/j.jclepro.2025.145644 Pavlovic, J., Tansini, A., Suarez, J., Fontaras, G., 2024. Influence of vehicle and battery ageing and driving modes on emissions and efficiency in Plug-in hybrid vehicles. Energy Conversion and Management: X 24, 100776. https://doi.org/10.1016/j.ecmx.2024.100776 Prakash, S., Bodisco, T.A., 2019. An investigation into the effect of road gradient and driving style on NOx emissions from a diesel vehicle driven on urban roads. Transportation Research Part D: Transport and Environment 72, 220–231. https://doi.org/10.1016/j.trd.2019.05.002 Prati, M.V., Costagliola, M.A., Giuzio, R., Corsetti, C., Beatrice, C., 2021. Emissions and energy consumption of a plug-in hybrid passenger car in Real Driving Emission (RDE) test. Transportation Engineering 4, 100069. https://doi.org/ 10.1016/j.treng.2021.100069 Selleri, T., Melas, A., Ferrarese, C., Franzetti, J., Giechaskiel, B., Suarez-Bertoa, R., 2022. Emissions from a Modern Euro 6d Diesel Plug-In Hybrid. Atmosphere 13, 1175. https://doi.org/10.3390/atmos13081175 Stogios, C., Kasraian, D., Roorda, M.J., Hatzopoulou, M., 2019. Simulating impacts of automated driving behavior and traffic conditions on vehicle emissions. Transportation Research Part D: Transport and Environment 76, 176–192. https://doi.org/10.1016/j.trd.2019.09.020 Takagi, N., Watanabe, T., Fushiki, S., Yamazaki, M., Asou, S., Nishimura, Y., 2010. Development of Exhaust and Evaporative Emissions Systems for Toyota THS II Plug-in Hybrid Electric Vehicle. SAE International Journal of Fuels and Lubricants 3, 406–413. https://doi.org/ 10.4271/2010-01-0831 Wang, Y., Su, S., Lai, Y., Luo, W., Hou, P., Lyu, T., Ge, Y., 2023. China 6 EGR gasoline vehicles without a GPF may struggle to meet the potential SPN10 limit. Environment International 181, 108306. https://doi.org/10.1016/j.envint.2023.108306 Wang, Y., Wang, J., Hao, C., Wang, X., Li, Q., Zhai, J., Ge, Y., Hao, L., Tan, J., 2021. Characteristics of instantaneous particle number (PN) emissions from hybrid electric vehicles under the real-world driving conditions. Fuel 286, 119466. https://doi.org/10.1016/j.fuel.2020.119466 Xie, B., Li, T., Liu, T., Chen, H., Li, H., Li, Y., 2024. Exploring high-emission driving behaviors of heavy-duty diesel vehicles based on engine principles under different road grade levels. Science of The Total Environment 951, 175443. https://doi.org/10.1016/j.scitotenv.2024.175443 Yang, Z., Ge, Y., Thomas, D., Wang, X., Su, S., Li, H., He, H., 2019. Real driving particle number (PN) emissions from China-6 compliant PFI and GDI hybrid electrical vehicles. Atmospheric Environment 199. https://doi.org/10.1016/j.atmosenv.2018.11.037 Yao, Y., Li, J., He, C., Chen, Y., Yu, H., Wang, J., Yang, N., Zhao, L., 2025. Research on particle emissions of light-duty hybrid electric vehicles in real driving. Atmospheric Pollution Research 16, 102332. https://doi.org/10.1016/j.apr.2024.102332 Zhu, F., Liu, Y., Lu, C., Huang, Q., Wang, C., 2024. Research on thermal energy management for PHEV based on NSGA-II optimization algorithm. Case Studies in Thermal Engineering 54, 104046. https://doi.org/10.1016/j.csite.2024.104046 Zhu, R., Wei, Y., He, L., Wang, M., Hu, J., Li, Z., Lai, Y., Su, S., 2024. Particulate matter emissions from light-duty gasoline vehicles under different ambient temperatures: Physical properties and chemical compositions. Science of The Total Environment 926, 171791. https://doi.org/10.1016/j.scitotenv.2024.171791 Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial.docx Cite Share Download PDF Status: Published Journal Publication published 03 Mar, 2026 Read the published version in Environmental Monitoring and Assessment → Version 1 posted Editorial decision: Revision requested 11 Oct, 2025 Reviews received at journal 08 Oct, 2025 Reviewers agreed at journal 18 Sep, 2025 Reviewers agreed at journal 16 Sep, 2025 Reviews received at journal 01 Sep, 2025 Reviewers agreed at journal 14 Aug, 2025 Reviewers invited by journal 14 Aug, 2025 Editor assigned by journal 11 Aug, 2025 Submission checks completed at journal 11 Aug, 2025 First submitted to journal 05 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7298392","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":502784431,"identity":"9e2da324-d8d2-497b-a0a3-289b9ecf82eb","order_by":0,"name":"Ruizhi Huang","email":"","orcid":"","institution":"Shenzhen Campus of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Ruizhi","middleName":"","lastName":"Huang","suffix":""},{"id":502784433,"identity":"3d2d6660-74ae-4ed3-b968-589f0430dae9","order_by":1,"name":"Yuzhuang Pian","email":"","orcid":"","institution":"Shenzhen Campus of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Yuzhuang","middleName":"","lastName":"Pian","suffix":""},{"id":502784435,"identity":"fedd3a09-cb0f-4a66-89e1-d1a314f2c299","order_by":2,"name":"Li Li","email":"","orcid":"","institution":"Shenzhen Campus of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Li","suffix":""},{"id":502784436,"identity":"f645804f-8c17-479a-95ae-90f32a89d16a","order_by":3,"name":"Yonghong Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYBACPmYGBmaGHzZAkrkBJmiAVwsbSAtjTxqIJFYLA8gWtsNAimgt7Dxm0gU856P52xkbGH+21SU2sDdvk2CouYPHYUAtMyxu5844zNjAzNt2OLGB51iZBMOxZ/i18PDczm0AaWFsO5DYIJFjJsHYcJiAFrZzufMPwxwm/4YoLQdyNwC1MPC2MQNt4SGkha3YmrcnOXcjUMthnnOHjdt40ootEo7h1sLPf3jjbZ4fdrnzzh8++PBHWZ1sP/vhjTc+1ODWAgQsEjDWAUY2SEwxJODTAIzJDwj2H/xKR8EoGAWjYGQCAC//S6icOSo3AAAAAElFTkSuQmCC","orcid":"","institution":"Shenzhen Campus of Sun Yat-sen University","correspondingAuthor":true,"prefix":"","firstName":"Yonghong","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-08-05 08:53:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7298392/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7298392/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10661-026-15124-4","type":"published","date":"2026-03-03T15:59:40+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89640019,"identity":"e9f2fcc2-b7ba-4da1-bea0-709c0ce53b6c","added_by":"auto","created_at":"2025-08-22 08:03:13","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":95693,"visible":true,"origin":"","legend":"\u003cp\u003ePEMS installation diagram: (a) vehicle; (b) PEMS equipment.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/87155923e2beb0b5b191d21e.jpg"},{"id":89640023,"identity":"9aa8718b-4a96-4e9d-8ec0-96b3674be69b","added_by":"auto","created_at":"2025-08-22 08:03:13","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":165845,"visible":true,"origin":"","legend":"\u003cp\u003eThe trace of the test route and trip dynamics validation diagram: (a) The trace of the test route (Altitude: 185-452 m); (b) RPA; (c) va\u003csub\u003epos\u003c/sub\u003e[95].\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/a3f8fceb277fb664cf93f43a.jpg"},{"id":89640020,"identity":"0ef2acc7-dc64-4655-a42b-44e1e77152de","added_by":"auto","created_at":"2025-08-22 08:03:13","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":76319,"visible":true,"origin":"","legend":"\u003cp\u003eCharacteristics during engine restart phase: (a) Distribution of engine-off interval and exhaust gas temperature at the start of each restart, which show mean value in parentheses; (b) PN emission distribution during engine restarts.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/d0b4af40ec4edf6555e385e6.jpg"},{"id":89640021,"identity":"b06e00b5-faec-4fc1-a6f3-cd6f133949f2","added_by":"auto","created_at":"2025-08-22 08:03:13","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":116915,"visible":true,"origin":"","legend":"\u003cp\u003eEngine parameters and PN rates at the test beginning, which marks emission peak rates in grey area and engine’s first start in red area: (a) Aggressive test1; (b) Calm test1.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/60e95c98423aa82fda54a529.jpg"},{"id":89640026,"identity":"4e2eac37-aa35-4929-9495-cd90655549cf","added_by":"auto","created_at":"2025-08-22 08:03:13","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":182652,"visible":true,"origin":"","legend":"\u003cp\u003eHES’s engine operating conditions (engine speed \u0026amp; engine load) and frequency distribution in different driving behaviors, which show mean values in parentheses.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/1aade5e284fa330832e8f1d5.jpg"},{"id":89641707,"identity":"6646c84f-49e4-4b8e-b0b0-f7407bf3fe7a","added_by":"auto","created_at":"2025-08-22 08:11:13","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":132737,"visible":true,"origin":"","legend":"\u003cp\u003eRadar plots of clusters, which are stratified by HES and high emission modes: (A) engine speed (rpm), (B) load (%), (C) acceleration (m/s²), (D) EGT (°C), (E) cold start proportion (%), (F) restart proportion (%).\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/8cd85bebceec99434e13a656.jpg"},{"id":104250729,"identity":"f761ca58-0fac-47bf-80e5-89f4258ff37a","added_by":"auto","created_at":"2026-03-09 16:06:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1580470,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/bacfd803-1b41-46e5-93b4-fa49aca803a9.pdf"},{"id":89641708,"identity":"9b16d9e0-10be-4085-9d09-7da4cc28c50b","added_by":"auto","created_at":"2025-08-22 08:11:13","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":309642,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7298392/v1/f429b9008ed7311da2cba309.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Driving behavior-adaptive particle emissions in plug-in hybrid electric vehicles: cumulative-transient characteristics and clustered patterns for real- world monitoring","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWith persistent environmental challenges and energy crises, the development of new energy vehicles has emerged as a pivotal strategy (Du et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). PHEVs mitigate range anxiety via dual-power systems compared to internal combustion engine vehicles (ICEVs) (Prati et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; F. Zhu et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), yet suffer frequent engine restarts that elevate PN emissions: under state of charge (SoC) sustain state in RDE, the number of restart for gasoline direct injection (GDI) PHEV turns to 140\u0026ndash;190/100 km and the engine-on total time is around 0.5, and the emission factor ranges from 3.0-3.7\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cb\u003e#\u003c/b\u003e/km, which is close to port fuel injection (PFI) ICEVs (Fan et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Karjalainen et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). To address lifecycle emissions, the PTI, a European regulatory framework primarily targeting diesel verification, is being adapted for PHEVs in progressive regions. However, current monitoring primarily focuses on idle conditions, which fails to accurately reflect dynamic emission events like cold starts and restarts that dominate real-world PN profiles (Bainschab et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Melas et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Miao et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Consequently, many studied focused on the PN characteristics of PHEV, which can instruct the design of dynamic scenarios for PTI, but most of them have identified SoC and battery ageing as significant factors to emission (Liu et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Pavlovic et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), few studies have considered the impact of driving behaviors. For ICEVs, driving behaviors create varying power demands in the same location, thereby influencing emissions (Fernandes et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Huang et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Xie et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024\u003c/span\u003e): Chandrashekar et al. found that aggressive driving were 18\u0026ndash;30% higher than the normal driving for CO, HC, CO2, and NOx emission rates (Chandrashekar et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). For PHEVs, driving behaviors are still significant, influencing the demand power and battery state (Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Few studies considered ACC behavior, which has shorter reacting times compared to human behaviors (Stogios et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and may have different characteristic. Besides, most studies have focused the effect of behaviors on emission factors and road types, neglecting instantaneous characteristic and specific segment, such as cold start, hot start (China Ⅵ regulation defines hot start when initial coolant temperature exceeds 40℃) and restart, which have significant impacts: 70% of total PN occur within 30 s after restart in hybrid electric vehicles (HEVs), and the emission factor of cold start is 4.2\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cb\u003e#\u003c/b\u003e/km, much higher than average value of total test (4.2\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{10}}\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cb\u003e#\u003c/b\u003e/km) (Karjalainen et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Park et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Yao et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eMoreover, the particle size of particulate matter is another focus of emission characteristics. Most studies concentrate on PN which larger than 23 nm in size, and could not present PN factually: for HEVs, particles sized between 10\u0026ndash;23 nm account for over 50% of total emissions (Wang et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). For GDI IECVs, particles sized between 6\u0026ndash;23 nm account for 43\u0026ndash;58% quantity of total emissions, and nucleation mode particles dominate emission of PFI vehicles (Hu et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Miao et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; R. Zhu et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOverall, current research overlooks two critical dimensions: behavior-specific mechanisms during high-emission transients, and emission characteristics of sub-23 nm particles that dominate nucleation-mode emissions. To address this, the study analyzes instantaneous and cumulative PN characteristics with smaller-sized particles across four typical driving behaviors, elucidating their impact mechanisms on real world PN emissions. Through cluster analysis, dominant high-emission patterns are classified and quantified, enabling the design of tailored dynamic monitoring scenarios, including next-generation PTI. The core contributions can be summarized as follows:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eThe effect of driving behaviors on cumulative PN emissions, including total test, cold start and engine restart by calculating emission factor and analyzing relevant parameters.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eThe effect of driving behaviors on instantaneous high emission by analyzing engine operations.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eEmploying the k-means clustering to aggregates high-emission patterns. It identifies metrics characterizing dominant high emission patterns across diverse driving behaviors, quantifies parametric thresholds of typical behaviors and enable behavior-adaptive monitoring scenarios.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eTo the best of our knowledge, this is the first work to evaluate the effect of driving behaviors on PN emission of a representative PHEV in real world conditions.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003eThis section first introduces the emission measurement system, test vehicle specifications, and experimental route design. Subsequently, we detail the calculation methodology for emission factors and analytical method for instantaneous PN rates.\u003c/p\u003e\u003cp\u003eEmission measurement\u003c/p\u003e\u003cp\u003eRDE of exhaust pollutants, including PN and CO, were measured by AVL M.O.V. E PEMS. This system comprised three core modules: gas analysis module (GAS PEMS AVL493), PN analysis module (PM PEMS AVL494) and exhaust flow meter (AVL495 EFM). As Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows, the PEMS system was installed at the back of the vehicle and an extension to the exhaust tailpipe was used to mount an exhaust flow meter (EFM). Exhaust gas temperature (EGT) sensor was installed in the extension of exhaust tailpipe, directly below the vehicle. Pretest and post-test calibration audits on PEMS were performed to verify the precision of PN emission. Other accessories included a GPS module and a OBD communication module, which could record the vehicle location information with 1 Hz sampling frequency, such as coolant temperature, engine speed and vehicle velocity.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTest vehicle and route\u003c/p\u003e\u003cp\u003eRDE tests were conducted in November 2023 in Chongqing, China. A commercially popular passenger car equipped with cooled exhaust gas recirculation (EGR) was selected for test. And the after-treatment device was a three-way catalyst (TWC) without a gasoline particulate filter (GPF). The vehicle was soaked under ambient conditions prior to testing, with auxiliary devices such as air conditioning turned off and total weight maintained consistently across all tests. Vehicle specifications were are detailed in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. Notably, due to high ambient temperatures and insufficient soaking time between consecutive tests, the initial coolant temperatures for calm, aggressive and ACC tests were higher.\u003c/p\u003e\u003cp\u003eAs Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows, the testing route almost covered the total urban area around the city. The testing duration of each test was between 6000\u0026ndash;11000 s, with a fixed testing distance of 114.9 km. As Table S2 shows, the initial SoC was maintained at approximately 50% under charge-sustaining conditions. Although the SOC at the end of each test exhibits variability, fuel-equivalent conversion analysis reveals that the battery energy is significantly lower than fuel consumption, accounting for 0\u0026ndash;11% of total energy expenditure, which is practically negligible. All tests were driven by different drivers, who were skilled and professional, and were instructed to keep their driving behaviors across the test. All driving behaviors were tested twice to ensure reproducibility.\u003c/p\u003e\u003cp\u003eTo qualify driving aggressiveness, some studies (Dhital et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Prakash and Bodisco, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) used relative positive acceleration (RPA) to distinguish driving behaviors quantitatively. Fernandes (Fernandes et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2024\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) used multiple vehicle operating variables to build a detecting and classifying model of driving behavior, and analyze the roundabouts by driving behaviors, operational performance and emissions. Thus, driving behaviors were divided based on RPA and va\u003csub\u003epos\u003c/sub\u003e[95] in this research. The va\u003csub\u003epos\u003c/sub\u003e[95] is the 95th percentile value of the second-by-second product of vehicle velocity and acceleration. As Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows, va\u003csub\u003epos\u003c/sub\u003e[95] of aggressive driving behavior on rural and motorway slightly exceeded the defined threshold while RPA values of calm driving behavior in urban areas were marginally below the threshold. This indicated a clear distinction between different driving behaviors, covering a wide range of operational conditions. Notably, the differences between ACC and aggressive tests were more pronounced than those between normal and calm tests, highlighting significant variations in emission characteristics between aggressive and conservative driving behaviors.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eEmission factor and energy consumption calculation\u003c/p\u003e\u003cp\u003eChina Ⅵ regulation defines cold and hot start from engine\u0026rsquo;s first start, and it excludes cold start and vehicle stop status (velocity\u0026thinsp;\u0026lt;\u0026thinsp;1 km/h) while calculating emission factor of total test. Thus, the emission factor \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{EF}_{i}\\:\\)\u003c/span\u003e\u003c/span\u003e(#/km) calculation can be divided into different phases:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\begin{array}{c}{EF}_{i}\\:\\text{=}\\frac{\\sum\\:_{\\text{j=0}}^{\\text{n}}{\\text{q}}_{\\text{j}}\\text{\u0026times;t}}{\\text{3600\u0026times;}\\sum\\:_{\\text{j=0}}^{\\text{n}}{\\text{v}}_{\\text{j}}\\text{\u0026times;t}}\\#\\left(1\\right)\\end{array}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn this formula, subscript \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i\\)\u003c/span\u003e\u003c/span\u003e represents \u003cem\u003ec, h, e\u003c/em\u003e for cold/hot start and total test respectively. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{q}_{j}\\:\\)\u003c/span\u003e\u003c/span\u003e(#/s) is the instantaneous PN emission rate during section \u003cem\u003ej\u003c/em\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{v}_{j}\\)\u003c/span\u003e\u003c/span\u003e (km/h) represents for driving velocity. By multiplying sampling interval \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:t\\)\u003c/span\u003e\u003c/span\u003e (1s) and summation, PN emission and total driving distance in each section can be calculated.\u003c/p\u003e\u003cp\u003eThe fuel-equivalent energy consumption \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:E\\)\u003c/span\u003e\u003c/span\u003e (kWh) is defined as:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:\\begin{array}{c}\\:E=\\frac{V\\times\\:{q}_{g}\\times\\:{\\eta\\:}_{en}}{{\\eta\\:}_{em}}\\#\\left(2\\right)\\end{array}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:V\\)\u003c/span\u003e\u003c/span\u003e represents the fuel consumption (L). \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{q}_{g\\:}(L/kWh)\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{n}_{en}\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{n}_{em}\\)\u003c/span\u003e\u003c/span\u003e denote the energy conversion factor, engine thermal efficiency and the transmission efficiency, with values of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:8.9\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:0.35\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:0.9\\)\u003c/span\u003e\u003c/span\u003e respectively, which are derived from calorific value conversion and experimental data calculations.\u003c/p\u003e\u003cp\u003eHigh emission sets (HES) on instantaneous PN rates\u003c/p\u003e\u003cp\u003eHigh PN emission sets (HES), as defined by Mera (Mera et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), are used to characterize instantaneous emission: instantaneous PN rates are sorted in descending order, and cumulative emissions are calculated accordingly. With increase in the amount of computational data, the desired percentages of total emissions (20%, 50%, 80%) are reached, which are HES-20, HES-50 and HES-80. Because of the decreasing order of HES calculation, the minimum instantaneous PN emission rate in HES-20, HES-50, HES-80 are observed as emission threshold, which is corresponding to the intersection of curve and HES lines. As shown in Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, HES-20, HES-50 and HES-80 represent the set of observations with the highest instantaneous PN emissions, which mean cumulative PN emissions are 20%, 50% and 80% of total. Besides, approximately 50% of the duration in each test had almost no emissions, and these periods are excluded from HES. For PFI HEV under RDE test, the PN thresholds of HES-20, HES-50, HES-80 are 27.36, 5.98, 1.14 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cb\u003e#\u003c/b\u003e/s, and the time thresholds are 0.12, 0.56, 2.43%, revealing a extremely unbalance in time series (Wang et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e"},{"header":"Results and discussion","content":"\u003cp\u003eCumulative PN characteristics in different driving behaviors\u003c/p\u003e\u003cp\u003eIn this section, we calculated segment-specific emissions and analyzed key parametric influences to examine cumulative PN characteristics under varying driving behaviors.\u003c/p\u003e\u003cp\u003eAs Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig. S2 show, emission factors throughout testing remained below China 6 regulatory limits (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:6\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cb\u003e#\u003c/b\u003e/km), outperforming both GDI PHEVs under CS state and PFI HEVs (Kontses et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Melas et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). However, emission varied substantially across driving behaviors: aggressive driving yielded the highest emissions, exceeding those of calm tests by a factor of two, while ACC achieved the lowest levels. Focusing on transient phases identified as critical PN sources in hybrid vehicles (Feinauer et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Selleri et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), cumulative PN is categorized into restart (PN generated within the first 15 seconds of each engine restart event), cold/hot start and other operations. To resolve temporal overlaps (restarts occurring during cold/hot start), Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e distinguishes restart contribution excluding cold/hot start phases and total restart contribution including overlaps, which indicate isolated restart impact and comprehensive engine-start burden. The results indicate that driving behaviors significantly influence the PN portion of restart and cold start, with restart (except hot/cold start) emissions comprising 19.9\u0026ndash;56.6% and cold start emissions ranging from 5.6\u0026ndash;49.9% of the total test. As critical emission sources and behavior-dependent variables, these segments necessitate phase-specific analysis to decode behavioral impacts.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTotal emissions, PN portion and parameters of different running segments.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDriving behavior and test set\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEmission factor of total test (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{10}}\\:\\)\u003c/span\u003e\u003c/span\u003e#/km)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eInitial engine coolant temperature (℃)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePN portion of cold/hot start (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eRestart contribution except hot/cold start phase (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eTotal restart contribution including overlaps (%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalm test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e49.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e29.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e76.3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalm test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e33.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e35.8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNormal test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e44.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e25.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e67.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNormal test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e41.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e28.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e59.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACC test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e36.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e39.8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACC test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e56.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e63.3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAggressive test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e19.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAggressive test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e5.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e26.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e32.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDriving behavior and test set\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNumber of restarts\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTotal test duration\u003c/p\u003e\u003cp\u003e(s)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTime portion of pure electric driving (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eEmission factor of cold/hot start (#/km)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eCoolant temperature rise rate until 70\u0026deg;C (\u0026deg;C/s)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalm test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10024\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e53.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e5.0\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.16\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalm test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8499\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e52.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.1\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.17\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNormal test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e106\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10628\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e46.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6.2\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.16\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNormal test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8570\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.3\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACC test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e108\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10977\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e29.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e8.0\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{1}\\text{0}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACC test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e110\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e44.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.6\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{1}\\text{0}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.23\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAggressive test 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e123\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e9600\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e51.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7.7\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{1}\\text{0}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAggressive test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6684\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e48.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.2\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe impact of restart on PN emission\u003c/p\u003e\u003cp\u003eRestart phases contribute dominantly (19.9\u0026ndash;56.6%) and behavior-sensitively to PN. Previous studies (Park et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) indicate that HEV restart strategies optimize energy efficiency and battery life by reducing unnecessary engine operation under low-load conditions (Choi et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Furthermore, restarts critically influence emissions through two thermal parameters: EGT, which determines catalyst conversion efficiency; engine-off interval, which governs catalyst cooling rate. Therefore, this section investigates the linkage between EGT and engine-off interval to examine how driving behaviors influence both restart strategy execution and resultant emissions.\u003c/p\u003e\u003cp\u003eAs Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e show, behavioral analysis reveals stark thermal divergences: aggressive driving shortened engine-off intervals by 50% versus calm driving while elevating restart EGT by 38\u0026ndash;64%, resulting in 68\u0026ndash;79% of restart PN concentrated within the initial 15s. ACC driving protocols extended engine-off intervals 39\u0026ndash;67% beyond normal, yielding comparable EGT but 21\u0026ndash;52\u0026deg;C lower per-restart emissions than calm driving.\u003c/p\u003e\u003cp\u003eThese thermal divergences fundamentally restructured PN profiles: aggressive driving elevated restart frequencies (111\u0026ndash;123 times during 6684\u0026ndash;9600 s of test duration) while it reduces per-event emission intensity: although frequent restarts raise cumulative PN counts, they depress average EGT per restart, thereby diminishing emission peaks. Crucially, under controlled SOC, high frequency restarts elevate temperature, enhancing catalyst conversion efficiency: the resultant emission reduction outweighs the particle increase from restart events. Consequently, aggression inversely correlates with restart-induced emissions. This redefines optimization priorities: extended engine-off intervals (\u0026gt;\u0026thinsp;5 min) causing critical catalyst cooling emerge as greater threats than restart frequency, demanding mitigation in next-generation energy management strategies.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eImpacts of engine cold/hot start\u003c/p\u003e\u003cp\u003eCold start phase contributes over 40% emission of total trips (Badshah et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Park et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), yet critically overlooks the significant influence of driving behaviors (3.1\u0026ndash;49.9%). In this section, this phase is dissected through two lenses: cold/hot start emission factor and contrasts and fixed engine warm-up strategy, including behavioral interference with thermal conditioning.\u003c/p\u003e\u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, cold start emission factors (0.77\u0026ndash;6.2\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e #/km) consistently exceeded hot start levels (0.46\u0026ndash;1.1\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{11}}\\)\u003c/span\u003e\u003c/span\u003e #/km). This disparity stems predominantly from sub-40\u0026deg;C catalyst inefficiency during cold start, where low EGT delay catalytic activation (Badshah et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). In contrast, hot start operations (initial coolant\u0026thinsp;\u0026gt;\u0026thinsp;40\u0026deg;C) maintain high catalyst efficiency, minimizing behavior-dependent emission variations.\u003c/p\u003e\u003cp\u003eAs Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows, vehicle tends to maintain a low engine load (\u0026lt;\u0026thinsp;50%) during the initial engine start in each test, which is also found in HEVs and ICEVs (Chambon et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Takagi et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Low engine load falls within the optimal range of exhaust gas recirculation (EGR) rates. It would reduce the burning temperature and promote more uniform combustion, which inhibits the formation of large carbonaceous agglomerates and decreases nucleation mode emissions (Guo et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Thus, this strategy helps maintain low emission rates while increasing TWC temperature, thereby reducing PN emission in subsequent engine restart. However, prioritizing fuel economy over emission reduction may compromise outcomes: low engine load reduced fuel atomization efficiency, and leads to incomplete combustion. Experimentally, this strategy was rigidly implemented, lasting precisely 60 seconds exclusively during the initial engine start of each test, irrespective of driving behavior, ambient conditions, or catalyst temperature. This phenomenon\u0026rsquo;s restriction to first-start events likely stems from insufficient engine off duration to trigger the strategy during subsequent cycles. Furthermore, calm and normal driving exhibit gradual power transitions and lower peak demands, slowing engine coolant temperature rise rate (0.14\u0026ndash;0.17\u0026deg;C/s) and number of restart and elevating PN emissions. As noted in prior research (Hu et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), emissions inherently represent a trade-off relation with fuel economy under energy management strategies. Nevertheless, strategy designers may have underestimated the substantial impact of driving behavior on particle emissions during cold start phases, thereby contributing to the occurrence of high emission events.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eInstantaneous PN threshold and time proportion of HES\u003c/p\u003e\u003cp\u003eAs Fig. S3 and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e show, the majority of emissions are generated within a small fraction of time. For normal and ACC tests, HES-20 are generated in less than 0.2% of test duration, and HES-80 are generated in less than 10% of test duration. Aggressive tests have a slightly higher time percentage, reaching 0.5% of duration for HES-20 and 20% for HES-80.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eThresholds of instantaneous PN rates and time proportion of HES in all tests.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"16\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c14\" colnum=\"14\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c15\" colnum=\"15\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c16\" colnum=\"16\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eDriving behavior and test set\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eCalm test 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c7\" namest=\"c4\"\u003e\u003cp\u003eNormal test 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c12\" namest=\"c8\"\u003e\u003cp\u003eACC test 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003eAggressive test 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"1\" nameend=\"c16\" namest=\"c16\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c13\" namest=\"c11\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c14\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHES-20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e84.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e98.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e9.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c13\" namest=\"c11\"\u003e\u003cp\u003e0.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e10.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003e0.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHES-50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e15.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003e0.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e16.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e2.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c13\" namest=\"c11\"\u003e\u003cp\u003e1.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e1.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003e4.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHES-80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003e3.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e0.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e0.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c13\" namest=\"c11\"\u003e\u003cp\u003e7.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e0.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003e20.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eDriving behavior and test set\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eCalm test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c8\" namest=\"c4\"\u003e\u003cp\u003eNormal test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"4\" nameend=\"c12\" namest=\"c9\"\u003e\u003cp\u003eACC test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003eAggressive test 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c16\" namest=\"c16\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{}\\text{\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003eThreshold\u003c/p\u003e\u003cp\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{10}}^{\\text{9}}\\text{\\:\\#/s}\\)\u003c/span\u003e\u003c/span\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003cp\u003e(%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHES-20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e85.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e41.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\u003cp\u003e24.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e30.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003e0.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHES-50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003e0.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e7.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e\u003cp\u003e1.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\u003cp\u003e4.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\u003cp\u003e0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e3.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003e2.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHES-80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003e4.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e0.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e\u003cp\u003e9.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\u003cp\u003e0.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\u003cp\u003e6.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u003cp\u003e0.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003e17.3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFurthermore, it is observed that for human driving behaviors (normal, aggressive, calm), the time proportion of HES exhibits a clear negative correlation with the PN rate threshold of HES. In normal test the PN rate threshold is 12.5-1013.3% higher than aggressive ones, and the time proportion is 82.1-1366.7% lower. Calm tests show similar trends to normal tests. With similar total emissions, Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e indicates aggressive tests do not have skyscraping emission rates compared with normal tests, and need more time to accumulate the same order of magnitude of emissions. The low cumulative emissions in ACC tests can be attributed to two factors: first, the PN thresholds for HES-20 and HES-80 are generally lower than those in aggressive tests while HES-50 thresholds are generally higher, suggesting a more uniform distribution of instantaneous PN rates. Besides, as shown in Fig. S3, when cumulative emissions of ACC tests reach 99.9%, the corresponding time percentages account for 84.0% and 82.5%. This indicates a significant proportion of time with almost no PN emissions, which is not entirely due to engine shutdown or electric driving. In summary, low PN emissions in ACC tests are a result of both lower peak rates and a higher time proportion of negligible emissions.\u003c/p\u003e\u003cp\u003eEngine speed and load impact on driving behaviors\u003c/p\u003e\u003cp\u003eBuilding upon earlier findings revealing consistent transient high-emission patterns across repeated tests under identical driving behaviors, this section consolidates tests with the same driving behavior to address the statistical limitations of HES-20 sampling.\u003c/p\u003e\u003cp\u003eEngine operating status is considered the most relevant parameter influencing PN. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e illustrates the engine operation, mean values, frequency distribution of HES, and mean engine operation values across the total test (except vehicle stop and electric driving state). Holistically, despite identical routes and engine strategies, compared to normal and calm driving, aggressive and ACC driving exhibit higher power demands, resulting in elevated engine speed and load distributions (Δ\u0026thinsp;=\u0026thinsp;80\u0026ndash;179 rpm, 0\u0026ndash;6% in mean value). Higher engine speed reduces fuel evaporation time, inducing localized incomplete combustion and increased PN emissions (Chen et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Concurrently, high engine load accelerates combustion deterioration, amplifying full-spectrum particle emissions, particularly for particles under 23 nm (Hu et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFor HES distributions, all driving behaviors show concentrated occurrence at around 2000 rpm. However, aggressive and conservative behaviors exhibit greater divergence in mean HES engine speeds (Δ\u0026thinsp;=\u0026thinsp;95-1094 rpm, 1\u0026ndash;33%), attributable to secondary emission peaks near 3000 rpm in aggressive and ACC driving. Thus, these behaviors concentrate high-emission zones under high engine operation (\u0026gt;\u0026thinsp;50% of PN spikes at \u0026gt;\u0026thinsp;2500 rpm and \u0026gt;\u0026thinsp;80% load), corroborating prior conclusions.\u003c/p\u003e\u003cp\u003eBesides, driving behavior emission differentials further correlate significantly with HES tendency of engine load and speed distribution. For calm and normal driving, HES-20 to HES-80 shows rising mean speed and load (194\u0026ndash;394 rpm, 9\u0026ndash;19%), indicating dominant emissions at lower speeds with secondary peaks at higher loads. This confirms cold start and restart emissions exceed those under stabilized high load conditions. For aggressive driving, HES-20 to HES-80 shows declining speed and load (252 rpm, 4%), demonstrating higher emissions during high load operation than others. For ACC driving, HES-20 to HES-80 exhibits a declining engine speed trajectory (54 rpm) coupled with rising load (9%), further evidencing higher emission peaks in high-speed regimes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eCurrent PTI for emission control prioritize idle conditions, which contributes\u0026thinsp;\u0026lt;\u0026thinsp;1% emission, misaligned with core high-emission engine states identified under RDE condition. This misalignment arises from PTI's static and low load testing versus real world dynamic high emission scenarios. Consequently, following the previous research of HES and high emission segments across diverse driving behaviors, these events are grouped by parameters to identify statistically robust high emission modes with defined parametric signatures. This enables quantitative linkage between emission patterns and driving behaviors, thereby facilitating behavior-differentiated dynamic inspection scenario design for emission-oriented vehicle monitoring, such as PTI.\u003c/p\u003e\u003cp\u003eClustering and behavior-adaptive monitoring for PN emissions\u003c/p\u003e\u003cp\u003eCluster analysis is recognized as an effective data processing method, widely adopted in emission characterization studies (Mehnatkesh et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This study applies k-means clustering to HES. Temporally continuous high-emission points are statistically characterized and consolidated as input data, with rotational engine speed, load, EGT, and acceleration serving as clustering features. The optimal cluster count is adaptively determined by maximizing the silhouette coefficient. Consistent with the earlier methodology addressing insufficient transient emission samples, tests under the same behaviors are aggregated. We statistically characterized the engine speed, load, velocity and acceleration at cluster centroids, along with the proportions of cold start and restart events within each cluster. As Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows, when multiple clusters share the same emission mode, HES level and driving behavior, their parameter values (90th percentile) are averaged for visualization, and the cluster count per group is annotated above each radar plot. The emission modes are divided into three modes, the threshold of restart and cold start is \u0026gt;\u0026thinsp;80% event; the threshold of high engine operation is engine speed\u0026thinsp;\u0026gt;\u0026thinsp;2800 rpm or load\u0026thinsp;\u0026gt;\u0026thinsp;80%.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e and Table S3 reveal behavioral emission patterns consistent with prior findings: 77.8% of clusters are classified as high engine operation in aggressive driving, thus it predominantly concentrates high emission clusters in high engine operation segments. Within high engine operating clusters, speed and load distributions converge across behaviors, yet aggressive driving exhibits significantly higher acceleration (0.5\u0026ndash;1.9 vs 0.0-0.7 m/s\u0026sup2;). This substantiates its pronounced impact on transient power output, necessitating dynamic tests with steeper power transitions and higher acceleration. ACC driving distributes clusters across all three modes, but 45% of clusters are high engine operation. The remaining clusters occur in HES-20 cold start, restart and HES-80 restart modes, explaining the inverted speed-load trajectory (Section 3.2.2). Compare to aggressive driving, calm and normal driving demonstrate inverse distribution patterns: all clusters are cold start dominant in HES-20, over 57% of clusters are restart-dominant in HES-50. For HES-80, clusters are split between high engine operation and restart modes. Besides, calm driving shows greater conservatism, retaining cold start clusters at HES-50. These results validate the efficacy of the classification thresholds and significant behavioral influences on high emission segments.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDriving behavior-adaptive focus on dominant emission modes and transient characteristics.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDriving behavior\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDominant emission mode\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFull process and transient high emission focus\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalm/ Normal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRestart (46.1% of HES events in calm, 50.9% of HES events in normal)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAverage start-up EGT \u0026minus;\u0026thinsp;40%,\u003c/p\u003e\u003cp\u003elower restart frequency and longer test duration vs. aggressive tests.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCalm/ Normal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCold start (1.6% of HES events in calm, 2.2% of HES events in normal)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eProhibit restarts in first 60s,\u003c/p\u003e\u003cp\u003ecoolant temperature rise rate\u0026thinsp;\u0026lt;\u0026thinsp;0.2\u0026deg;C/s,\u003c/p\u003e\u003cp\u003epay attention to the sudden increase in engine speed load after 60 seconds or the second restart.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRestart (36.6% of HES events)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAverage start-up EGT \u0026minus;\u0026thinsp;30% compare to aggressive tests.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eACC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHigh engine operation (59.1% of HES events)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEngine speed threshold\u0026thinsp;\u0026gt;\u0026thinsp;2800 rpm,\u003c/p\u003e\u003cp\u003eengine load threshold\u0026thinsp;\u0026gt;\u0026thinsp;80%,\u003c/p\u003e\u003cp\u003emedium acceleration.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAggressive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHigh engine operation (58.5% of HES events)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEngine speed threshold\u0026thinsp;\u0026gt;\u0026thinsp;2800 rpm,\u003c/p\u003e\u003cp\u003eengine load threshold\u0026thinsp;\u0026gt;\u0026thinsp;80%,\u003c/p\u003e\u003cp\u003eaverage acceleration\u0026thinsp;+\u0026thinsp;50% vs. ACC tests.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eAs Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows, based on the above conclusions and analysis of high emission modes and HES events, the high emission parameter characteristics and emission profiles of PHEVs under different driving behaviors are summarized. Driving behaviors systematically reconfigure the parametric signatures and modal dominance hierarchies of emission-critical states, necessitating three key adaptations in dynamic monitoring design: temporal allocation must prioritize high engine operation for aggressive driving (58.5% of HES events) versus restart phases for conservative driving (46.1/50.9%); parameter scaling should reflect behavior-specific extremes, expanding acceleration ranges from 0.5 to 1.9 m/s\u0026sup2; to capture aggressive transients. Critically, while behavioral variations reprioritize monitoring focus across modes, no emission critical state loses relevance. Thus, dynamic scenarios must retain but reweight all modal components based on driving behavior.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eA representative hybrid system-equipped PHEV (EGR-enabled, GPF-free) was analyzed for cumulative and instantaneous PN characteristics across various driving behaviors under real world conditions, with operational parameters examined as influencing factors. Cumulative PN characteristics were evaluated across three specific segments: cold/hot start and restart. Normal and calm driving have higher PN portion from restart (19.9\u0026ndash;56.6%) and cold start (41.8\u0026ndash;49.9%) than aggressive behaviors. This is attributed to lower and smoother power demands, resulting in longer engine-off interval (53.2\u0026ndash;74.1 s average) and lower initial restart EGT (62.0-70.6\u0026deg;C average). Thus, it is proved that extending engine-off intervals causing critical catalyst cooling emerge as greater threats than restart frequency for PN emission. ACC driving achieve the lowest total emissions because of significantly extended zero-emission durations (16.0-17.5%). While cold start engine strategies balance fuel economy and emission control through low-load operation (\u0026lt;\u0026thinsp;50%) to promote uniform combustion, the impact of driving behavior on thermal conditioning is overlook: aggressive electric assist during calm and normal driving slows coolant warming (0.14\u0026ndash;0.17\u0026deg;C/s) and decrease the number of restart, exacerbating PN emission variance by up to 8-fold (7.7\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{10}}\\)\u003c/span\u003e\u003c/span\u003e-6.2\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{\u0026times;}{\\text{10}}^{\\text{1}\\text{1}}\\:\\)\u003c/span\u003e\u003c/span\u003e#/km) across tests. Overall, the emission reduction potential of driving behavior can be quantitatively inferred. That is to say, the transition from radical to conservative driving can reduce PN by 54% and ACC by further reducing emission duration by 17%.\u003c/p\u003e\u003cp\u003eFor instantaneous emission, HES analysis reveals that over 80% of PN emissions concentrate in less than 10% of duration, with emission thresholds exhibiting profound behavioral divergence: aggressive driving requires 82\u0026ndash;1366% lower instantaneous PN rates to trigger HES events compared to normal driving, yet sustains high emission phases 20.4% longer (HES-80). This volatility originates from behavioral modulation of engine operating states: aggressive driving concentrated 77.8% of high-emission events at high load and speed regimes (\u0026gt;\u0026thinsp;2800 rpm or \u0026gt;\u0026thinsp;80% load) while calm and normal driving prioritized cold start and restart contributions. These parametric divergences necessitate shifting focus from emission levels to engine dynamics. Cluster analysis (k-means) resolves this by categorizing 85% of clusters into three high emission modes, while exposing behavioral nuances within modes: identical high-load clusters show higher acceleration (50%) in aggressive driving, and cold start clusters exhibit slower coolant warming in conservative driving.\u003c/p\u003e\u003cp\u003eTherefore, for behavior-adaptive monitoring, aggressive driving demands high proportion of high engine operation segments with higher acceleration (0.5\u0026ndash;1.9 m/s\u0026sup2;), triggered at \u0026gt;\u0026thinsp;2800 rpm and 80% load thresholds to capture their extreme emissions. Conservative driving requires extended restart and cold start segments, monitoring the sudden increase in engine speed load after 60 seconds or the second restart; ACC driving necessitates hybrid scenarios balancing restart transients (EGT\u0026thinsp;+\u0026thinsp;10% compensation compare to conservative ones) and sustained high engine operation.\u003c/p\u003e\u003cp\u003eThe shift in emission monitoring (such as PTI) from static to dynamic paradigms is evident, and it would be better to incorporate driving behavior variability into emission regulations and certification protocols to ensure real-world compliance. Behavioral variations reprioritize monitoring focus across modes, but no emission critical mode loses relevance. Thus, dynamic scenarios must retain but reweight all modal components by temporal allocation and parameter scaling based on driving behavior.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis work was supported by the Fundamental Research Key Program of Shenzhen (JCYJ20241202130016022) and National Natural Science Foundation of China (41975165).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eRuizhi Huang: Methodology, Formal analysis, Writing-original draft, Visualization; Yuzhuang Pian: Investigation, Data curation, Writing-Review \u0026amp; editing; Li Li: Project administration, Conceptualization; Yonghong Liu: Supervision, Funding acquisition, Resources.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBadshah, H., Kittelson, D., Northrop, W., 2016. Particle Emissions from Light-Duty Vehicles during Cold-Cold Start. \u003cem\u003eSAE Int. J. Engine\u003c/em\u003es 9, 1775\u0026ndash;1785. https://doi.org/10.4271/2016-01-0997\u003c/li\u003e\n\u003cli\u003eBainschab, M., Schriefl, M.A., Bergmann, A., 2020. Particle number measurements within periodic technical inspections: A first quantitative assessment of the influence of size distributions and the fleet emission reduction. \u003cem\u003eAtmospheric Environment: X\u003c/em\u003e 8, 100095. https://doi.org/10.1016/j.aeaoa.2020.100095\u003c/li\u003e\n\u003cli\u003eChambon, P., Deter, D., Irick, D.K., Smith, D.E., 2013. PHEV Cold Start Emissions Management. \u003cem\u003eSAE International Journal of Alternative Powertrains\u003c/em\u003e 2, 252\u0026ndash;260. https://doi.org/ 10.4271/2013-01-0358\u003c/li\u003e\n\u003cli\u003eChandrashekar, C., Rawat, R.S., Chatterjee, P., Pawar, D.S., 2023. Evaluating the real-world emissions of diesel passenger Car in Indian heterogeneous traffic. \u003cem\u003eEnviron Monit Assess\u003c/em\u003e 195, 1248. https://doi.org/10.1007/s10661-023-11658-z\u003c/li\u003e\n\u003cli\u003eChen, L., Liang, Z., Zhang, X., Shuai, S., 2017. Characterizing particulate matter emissions from GDI and PFI vehicles under transient and cold start conditions. \u003cem\u003eFuel\u003c/em\u003e 189, 131\u0026ndash;140. https://doi.org/10.1016/j.fuel.2016.10.055\u003c/li\u003e\n\u003cli\u003eChoi, Y., Yi, H., Oh, Y., Park, S., 2021. Effects of engine restart strategy on particle number emissions from a hybrid electric vehicle equipped with a gasoline direct injection engine. \u003cem\u003eAtmospheric Environment\u003c/em\u003e 253, 118359. https://doi.org/10.1016/j.atmosenv.2021.118359\u003c/li\u003e\n\u003cli\u003eDhital, N.B., Wang, S.-X., Lee, C.-H., Su, J., Tsai, M.-Y., Jhou, Y.-J., Yang, H.-H., 2021. Effects of driving behavior on real-world emissions of particulate matter, gaseous pollutants and particle-bound PAHs for diesel trucks. \u003cem\u003eEnvironmental Pollution\u003c/em\u003e 286, 117292. https://doi.org/10.1016/j.envpol.2021.117292\u003c/li\u003e\n\u003cli\u003eDu, J., Ouyang, M., Chen, J., 2017. Prospects for Chinese electric vehicle technologies in 2016\u0026ndash;2020: Ambition and rationality. \u003cem\u003eEnergy\u003c/em\u003e 120, 584\u0026ndash;596. https://doi.org/10.1016/j.energy.2016.11.114\u003c/li\u003e\n\u003cli\u003eFan, Q., Wang, Y., Xiao, J., Wang, Z., Li, W., Jia, T., Zheng, B., Taylor, R.I., 2020. Effect of Oil Viscosity and Driving Mode on Oil Dilution and Transient Emissions Including Particle Number in Plug-In Hybrid Electric Vehicle. \u003cem\u003eSAE Technical Paper Series\u003c/em\u003e. https://doi.org/ 10.4271/2020-01-0362\u003c/li\u003e\n\u003cli\u003eFeinauer, M., Ehrenberger, S., Epple, F., Schripp, T., Grein, T., 2022. Investigating Particulate and Nitrogen Oxides Emissions of a Plug-In Hybrid Electric Vehicle for a Real-World Driving Scenario. \u003cem\u003eApplied Sciences\u003c/em\u003e 12, 1404. https://doi.org/10.3390/app12031404\u003c/li\u003e\n\u003cli\u003eFernandes, P., Ferreira, E., Macedo, E., Coelho, M.C., 2024. Unraveling roundabout dynamics: Analysis of driving behavior, vehicle performance, and exhaust emissions. \u003cem\u003eTransportation Research Part D: Transport and Environment\u003c/em\u003e 133, 104308. https://doi.org/10.1016/j.trd.2024.104308\u003c/li\u003e\n\u003cli\u003eFernandes, P., Macedo, E., Tom\u0026aacute;s, R., Coelho, M.C., 2023. Hybrid electric vehicle data-driven insights on hot-stabilized exhaust emissions and driving volatility. \u003cem\u003eInternational Journal of Sustainable Transportation\u003c/em\u003e 18, 84\u0026ndash;102. https://doi.org/10.1080/15568318.2023.2219629\u003c/li\u003e\n\u003cli\u003eFernandes, P., Tom\u0026aacute;s, R., Ferreira, E., Bahmankhah, B., Coelho, M.C., 2021. Driving aggressiveness in hybrid electric vehicles: Assessing the impact of driving volatility on emission rates. \u003cem\u003eApplied Energy\u003c/em\u003e 284, 116250. https://doi.org/10.1016/j.apenergy.2020.116250\u003c/li\u003e\n\u003cli\u003eGuo, Z., Wang, Boyuan, Liu, S., Zhang, Z., Wang, Buyu, Chang, C.-T., Wang, P., He, X., Sun, X., Shuai, S., 2023. Experimental investigation on emission characteristics of high reactivity gasoline compression ignition with different EGR and injection pressure strategies. \u003cem\u003eFuel\u003c/em\u003e 332, 126176. https://doi.org/10.1016/j.fuel.2022.126176\u003c/li\u003e\n\u003cli\u003eHu, R., Chen, X., Li, L., Kong, F., Liu, Y., 2024. Exhaust emissions and energy conversion of hybrid and conventional CNG buses. \u003cem\u003eTransportation Research Part D: Transport and Environment\u003c/em\u003e 135, 104405. https://doi.org/10.1016/j.trd.2024.104405\u003c/li\u003e\n\u003cli\u003eHu, Z., Lu, Z., Song, B., Quan, Y., 2020. Impact of test cycle on mass, number and particle size distribution of particulates emitted from gasoline direct injection vehicles. \u003cem\u003eThe Science of the total environment\u003c/em\u003e 143128. https://doi.org/ 10.1016/j.scitotenv.2020.143128\u003c/li\u003e\n\u003cli\u003eHu, Z., Xu, Y., Wang, Z., Zhang, H., Tan, P., Lou, D., 2023. An experimental study on particle number, micromorphology and nanostructure characteristics of particulate matter from a China Ⅵ gasoline direct injection engine. \u003cem\u003eAtmospheric Environment: X\u003c/em\u003e. https://doi.org/ 10.1016/j.aeaoa.2023.100211\u003c/li\u003e\n\u003cli\u003eHuang, R., Ni, J., Cheng, Z.X., Wang, Q., Shi, X., Yao, X., 2021. Assessing the effects of ethanol additive and driving behaviors on fuel economy, particle number, and gaseous emissions of a GDI vehicle under real driving conditions. \u003cem\u003eFuel\u003c/em\u003e 306, 121642. https://doi.org/10.1016/j.envadv.2023.100454\u003c/li\u003e\n\u003cli\u003eKarjalainen, P., Leinonen, V., Olin, M., Vesisenaho, K., Marjanen, P., J\u0026auml;rvinen, A., Simonen, P., Markkula, L., Kuuluvainen, H., Keskinen, J., Mikkonen, S., 2024. Real-world emissions of nanoparticles, particulate mass and black carbon from a plug-in hybrid vehicle compared to conventional gasoline vehicles. \u003cem\u003eEnvironmental Advances\u003c/em\u003e 15, 100454. https://doi.org/10.1016/j.envadv.2023.100454\u003c/li\u003e\n\u003cli\u003eKontses, A., Triantafyllopoulos, G., Ntziachristos, L., Samaras, Z., 2020. Particle number (PN) emissions from gasoline, diesel, LPG, CNG and hybrid-electric light-duty vehicles under real-world driving conditions. \u003cem\u003eAtmospheric Environment\u003c/em\u003e 222, 117126.https://doi.org/10.1016/j.atmosenv.2019.117126\u003c/li\u003e\n\u003cli\u003eLi, B., Chen, Z., Zhu, X., Zhang, Z., Peng, Z., Zhao, H., He, H., 2024. Assessment of eco-driving strategies on carbon emissions for hybrid vehicles through portable emissions measurement systems. \u003cem\u003eAtmospheric Pollution Research\u003c/em\u003e 102365. https://doi.org/10.1016/j.apr.2024.102365\u003c/li\u003e\n\u003cli\u003eLiu, Y., Li, C., Boger, T., Feng, X., Li, W., Chen, X., 2024. RDE PN Emission Challenges for a China 6 PHEV. \u003cem\u003eSAE Technical Paper Series.\u003c/em\u003e https://doi.org/ 10.4271/2024-01-2386\u003c/li\u003e\n\u003cli\u003eMehnatkesh, H., Gordon, D., Koch, C.R., 2025. Dynamic emission analysis of a hydrogen/diesel dual-fuel engine using clustering method. \u003cem\u003eInternational Journal of Hydrogen Energy\u003c/em\u003e 136, 371\u0026ndash;382. https://doi.org/10.1016/j.ijhydene.2025.04.385\u003c/li\u003e\n\u003cli\u003eMelas, A., Selleri, T., Franzetti, J., Ferrarese, C., Suarez-Bertoa, R., Giechaskiel, B., 2022. On-Road and Laboratory Emissions from Three Gasoline Plug-In Hybrid Vehicles-Part 2: Solid Particle Number Emissions. \u003cem\u003eEnergies\u003c/em\u003e 15, 5266. https://doi.org/10.3390/en15145266\u003c/li\u003e\n\u003cli\u003eMelas, A., Vasilatou, K., Suarez-Bertoa, R., Giechaskiel, B., 2023. Laboratory measurements with solid particle number instruments designed for periodic technical inspection (PTI) of vehicles. \u003cem\u003eMeasurement\u003c/em\u003e 215, 112839. https://doi.org/10.1016/j.measurement.2023.112839\u003c/li\u003e\n\u003cli\u003eMera, Z., Fonseca, N., L\u0026oacute;pez, J.-M., Casanova, J., 2019. Analysis of the high instantaneous NOx emissions from Euro 6 diesel passenger cars under real driving conditions. \u003cem\u003eApplied Energy\u003c/em\u003e 242, 1074\u0026ndash;1089. https://doi.org/10.1016/j.apenergy.2019.03.120\u003c/li\u003e\n\u003cli\u003eMiao, X., Zhang, X., Wang, C., Li, J., Zhao, J., Qu, L., Liu, Y., Qi, S., Li, H., Fu, M., Jin, T., 2024. Effects of fuel and driving conditions on particle number emissions of China-VI gasoline vehicles: based on corrections to test results. \u003cem\u003eEnviron Monit Assess\u003c/em\u003e 196, 591. https://doi.org/10.1007/s10661-024-12756-2\u003c/li\u003e\n\u003cli\u003ePark, J., Kim, H., Kim, Y., Park, S., 2025. Real-world particle number emissions from hybrid electric vehicles with port fuel injection and dual injection (MPI-GDI) systems. \u003cem\u003eJournal of Cleaner Production\u003c/em\u003e 510, 145644. https://doi.org/10.1016/j.jclepro.2025.145644\u003c/li\u003e\n\u003cli\u003ePavlovic, J., Tansini, A., Suarez, J., Fontaras, G., 2024. Influence of vehicle and battery ageing and driving modes on emissions and efficiency in Plug-in hybrid vehicles. \u003cem\u003eEnergy Conversion and Management: X\u003c/em\u003e 24, 100776. https://doi.org/10.1016/j.ecmx.2024.100776\u003c/li\u003e\n\u003cli\u003ePrakash, S., Bodisco, T.A., 2019. An investigation into the effect of road gradient and driving style on NOx emissions from a diesel vehicle driven on urban roads. \u003cem\u003eTransportation Research Part D: Transport and Environment\u003c/em\u003e 72, 220\u0026ndash;231. https://doi.org/10.1016/j.trd.2019.05.002\u003c/li\u003e\n\u003cli\u003ePrati, M.V., Costagliola, M.A., Giuzio, R., Corsetti, C., Beatrice, C., 2021. Emissions and energy consumption of a plug-in hybrid passenger car in Real Driving Emission (RDE) test. \u003cem\u003eTransportation Engineering\u003c/em\u003e 4, 100069. https://doi.org/ 10.1016/j.treng.2021.100069\u003c/li\u003e\n\u003cli\u003eSelleri, T., Melas, A., Ferrarese, C., Franzetti, J., Giechaskiel, B., Suarez-Bertoa, R., 2022. Emissions from a Modern Euro 6d Diesel Plug-In Hybrid.\u003cem\u003e Atmosphere\u003c/em\u003e 13, 1175. https://doi.org/10.3390/atmos13081175\u003c/li\u003e\n\u003cli\u003eStogios, C., Kasraian, D., Roorda, M.J., Hatzopoulou, M., 2019. Simulating impacts of automated driving behavior and traffic conditions on vehicle emissions. \u003cem\u003eTransportation Research Part D: Transport and Environment\u003c/em\u003e 76, 176\u0026ndash;192. https://doi.org/10.1016/j.trd.2019.09.020\u003c/li\u003e\n\u003cli\u003eTakagi, N., Watanabe, T., Fushiki, S., Yamazaki, M., Asou, S., Nishimura, Y., 2010. Development of Exhaust and Evaporative Emissions Systems for Toyota THS II Plug-in Hybrid Electric Vehicle. \u003cem\u003eSAE International Journal of Fuels and Lubricants\u003c/em\u003e 3, 406\u0026ndash;413. https://doi.org/ 10.4271/2010-01-0831\u003c/li\u003e\n\u003cli\u003eWang, Y., Su, S., Lai, Y., Luo, W., Hou, P., Lyu, T., Ge, Y., 2023. China 6 EGR gasoline vehicles without a GPF may struggle to meet the potential SPN10 limit. \u003cem\u003eEnvironment International\u003c/em\u003e 181, 108306. https://doi.org/10.1016/j.envint.2023.108306\u003c/li\u003e\n\u003cli\u003eWang, Y., Wang, J., Hao, C., Wang, X., Li, Q., Zhai, J., Ge, Y., Hao, L., Tan, J., 2021. Characteristics of instantaneous particle number (PN) emissions from hybrid electric vehicles under the real-world driving conditions. \u003cem\u003eFuel\u003c/em\u003e 286, 119466. https://doi.org/10.1016/j.fuel.2020.119466\u003c/li\u003e\n\u003cli\u003eXie, B., Li, T., Liu, T., Chen, H., Li, H., Li, Y., 2024. Exploring high-emission driving behaviors of heavy-duty diesel vehicles based on engine principles under different road grade levels.\u003cem\u003e Science of The Total Environment \u003c/em\u003e951, 175443. https://doi.org/10.1016/j.scitotenv.2024.175443\u003c/li\u003e\n\u003cli\u003eYang, Z., Ge, Y., Thomas, D., Wang, X., Su, S., Li, H., He, H., 2019. Real driving particle number (PN) emissions from China-6 compliant PFI and GDI hybrid electrical vehicles. \u003cem\u003eAtmospheric Environment\u003c/em\u003e 199. https://doi.org/10.1016/j.atmosenv.2018.11.037\u003c/li\u003e\n\u003cli\u003eYao, Y., Li, J., He, C., Chen, Y., Yu, H., Wang, J., Yang, N., Zhao, L., 2025. Research on particle emissions of light-duty hybrid electric vehicles in real driving. \u003cem\u003eAtmospheric Pollution Research 16,\u003c/em\u003e 102332. https://doi.org/10.1016/j.apr.2024.102332\u003c/li\u003e\n\u003cli\u003eZhu, F., Liu, Y., Lu, C., Huang, Q., Wang, C., 2024. Research on thermal energy management for PHEV based on NSGA-II optimization algorithm. \u003cem\u003eCase Studies in Thermal Engineering\u003c/em\u003e 54, 104046. https://doi.org/10.1016/j.csite.2024.104046\u003c/li\u003e\n\u003cli\u003eZhu, R., Wei, Y., He, L., Wang, M., Hu, J., Li, Z., Lai, Y., Su, S., 2024. Particulate matter emissions from light-duty gasoline vehicles under different ambient temperatures: Physical properties and chemical compositions. \u003cem\u003eScience of The Total Environment\u003c/em\u003e 926, 171791. https://doi.org/10.1016/j.scitotenv.2024.171791\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"environmental-monitoring-and-assessment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"emas","sideBox":"Learn more about [Environmental Monitoring and Assessment](http://link.springer.com/journal/10661)","snPcode":"10661","submissionUrl":"https://submission.nature.com/new-submission/10661/3","title":"Environmental Monitoring and Assessment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"plug-in hybrid electric vehicle (PHEV), particle number (PN) emission, driving behavior, real driving emission (RDE)","lastPublishedDoi":"10.21203/rs.3.rs-7298392/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7298392/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRegulatory gaps in restart and cold/hot start emissions overlooked by current periodic technical inspection (PTI), and driving behaviors significantly impact plug-in hybrid electric vehicle (PHEV) particle number (PN) emissions under real driving conditions. Using portable emissions measurement systems (PEMS), this study building cumulative PN emissions across key segments (cold-start, restart) and instantaneous high-emission events across four distinct behaviors. Key findings reveal that calm and normal driving elevate cold-start PN (up to 6.2\u0026times;10\u0026sup1;\u0026sup1; #/km) due to prolonged engine-off intervals and slow warm-up. Aggressive driving\u0026rsquo;s frequent restarts yield lower per-event emissions owing to thermal advantages. Adaptive cruise control (ACC) minimizes total PN by combining thermally efficient engine operation with extended zero-emission phases (16\u0026ndash;17% duration). Crucially, instantaneous high-emission analysis shows\u0026thinsp;\u0026gt;\u0026thinsp;80% of PN concentrates in \u0026lt;\u0026thinsp;10% of driving duration, with emission thresholds varying dramatically (82-1366%) across behaviors\u0026mdash;primarily due to divergent dominant modes favored by each behavior. To quantify these behavior-specific modes and their parametric signatures, k-means clustering was applied, and found distinct behavioral associations: aggressive driving predominantly linked to high-load/high-rpm operation (\u0026gt;\u0026thinsp;2800 rpm or \u0026gt;\u0026thinsp;80% load), while calm/normal driving elevates cold-start and restart contributions. Consequently, real-world emission monitoring necessitates behavior-adaptive dynamic scenarios, tailoring test focus and parametric design informed by clustered thresholds.\u003c/p\u003e","manuscriptTitle":"Driving behavior-adaptive particle emissions in plug-in hybrid electric vehicles: cumulative-transient characteristics and clustered patterns for real- world monitoring","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-22 08:03:08","doi":"10.21203/rs.3.rs-7298392/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-11T13:01:48+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-08T13:43:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"54722078104686897927172699285292142407","date":"2025-09-18T06:59:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"137319930999323032184494757654229493446","date":"2025-09-16T08:35:00+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-01T17:20:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"160740797522448420844179789559960732170","date":"2025-08-14T15:22:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-14T12:38:58+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-11T23:16:56+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-11T23:16:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Environmental Monitoring and Assessment","date":"2025-08-05T08:39:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"environmental-monitoring-and-assessment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"emas","sideBox":"Learn more about [Environmental Monitoring and Assessment](http://link.springer.com/journal/10661)","snPcode":"10661","submissionUrl":"https://submission.nature.com/new-submission/10661/3","title":"Environmental Monitoring and Assessment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"734b0b47-410b-4f80-a5d3-6833f060f2a7","owner":[],"postedDate":"August 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-09T16:03:29+00:00","versionOfRecord":{"articleIdentity":"rs-7298392","link":"https://doi.org/10.1007/s10661-026-15124-4","journal":{"identity":"environmental-monitoring-and-assessment","isVorOnly":false,"title":"Environmental Monitoring and Assessment"},"publishedOn":"2026-03-03 15:59:40","publishedOnDateReadable":"March 3rd, 2026"},"versionCreatedAt":"2025-08-22 08:03:08","video":"","vorDoi":"10.1007/s10661-026-15124-4","vorDoiUrl":"https://doi.org/10.1007/s10661-026-15124-4","workflowStages":[]},"version":"v1","identity":"rs-7298392","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7298392","identity":"rs-7298392","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}