Machine Learning assisted Prediction of Heat Transfer Coefficients during Quenching

In this study, the Heat Transfer Coefficients (HTC) occurring during immersion quenching are predicted using a machine learning regression technique. This paper describes a statistical analysis of HTC by developing an artificial neural network-based machine learning model.

The effects of variation in the quecnhant's temperature, initial temperature and characteristics of measured cooling curves have been analyzed. The ANN has been trained on data acquisited during several types (ie: oil, polymer, spray, etc) and conditions (agitation, temperature, ageing, etc) of liquid quenchants. An Artificial Neural Network (ANN)model is used for regression analysis to predict the HTC in terms of temperature signals recorded, and the results showed high prediction accuracies. The applied ANN model seems to be robust and precise, and could be used by Heat Treatment design engineers for predicting the outputs of hardening processes.