Low-cycle fatigue Behavior Modeling of Similar and Dissimilar Carbon Steel under Rotary Friction Welding Effect using Adaptive Neuro-Fuzzy Inference System (ANFIS)

Rotary Friction Welding (RFW) plays a crucial role in manufacturing components for automotive and marine applications. However, the low-cycle fatigue life of dissimilar welded joints made of carbon steel alloys, specifically C35 and C45, remains insufficiently understood. This study aims to bridge this gap by evaluating how RFW parameters affect the fatigue life of these materials, using both experimental and modelling approaches. A series of axial low-cycle fatigue tests were conducted on base metals and RFW specimens under varying friction pressures, revealing a direct correlation between friction pressure and fatigue strength coefficient. The fatigue life was initially modelled using the Coffin-Manson equation, validated with established methods, and subsequently refined using an Adaptive Neuro-Fuzzy Inference System (ANFIS). The ANFIS model was developed to predict fatigue life under different stress and strain conditions, providing enhanced prediction accuracy compared to traditional empirical models. Results demonstrated that increased friction pressure significantly enhances fatigue life and strengthens weld interfaces. Detailed microstructural analysis further supported the observed improvements in fatigue performance. This research provides comprehensive insights into optimizing RFW parameters to improve the fatigue performance of both similar and dissimilar carbon steel joints. By integrating empirical, experimental, and ANFIS-based modelling, the findings offer practical guidelines for selecting optimal welding conditions in industrial applications, potentially reducing costs associated with extensive fatigue testing and improving component longevity.