Prediction of peak temperature value in friction lap welding of aluminium alloy 7475 and PPS polymer hybrid joint using machine learning approaches

•Machine learning models were employed to predict the temperature value.•Both SVM and GPR models with PUK kernel achieved high accuracy for prediction.•SVM-PUK approach should be prioritized for temperature prediction during FLW. In the automotive and aerospace sectors, vehicle weight has been a major concern for the past decades. Incorporating metal-polymer hybrid structures reduces weight without compromising structural performance. In this work, Friction Lap Welding (FLW) technique was used to join aluminium alloy 7475 with PPS polymer. To conduct the FLW experiments, a design matrix was developed with the help of Response surface methodology. For each experimental run, the peak temperature value was recorded. The possibility of machine learning algorithms for predicting peak temperature during joining has been assessed. The maximum correlation coefficient (i.e., 0.999) and minimum root mean square error (i.e., 0.068) were observed for the support vector machining (SVM) model with the PUK kernel. So, SVM-PUK approach was determined to be the most efficient.