Estimation of Safe Operating Area for EV Battery

Abstract In recent times, electric vehicles (EVs) have experienced battery pack failures dueto various factors such as short circuits, thermal imbalances, fires, and explosions.This research paper employs a machine-learning approach to estimate lithium-ion batteries safe operating area (SOA). It explores the factors that influencebattery safety levels and discusses the safe operating area for charging and dis-charging based on real-time data. The SOA concept is a critical consideration inthe design, operation, and management of EV batteries, as it defines the limitswithin which the battery can operate safely without compromising performanceor endangering the vehicle. The significance of SOA analysis for EV batteriesand its implications on battery design, thermal management, charging strate-gies, and overall vehicle safety. This paper specifically addresses the estimationof the safe operation of a battery using ML in relation to voltage and temper-ature with respect to time. It emphasises the importance of understanding theinterplay between these parameters to establish safe operational limits.