Prediction of age hardening parameters for 17-4PH stainless steel by artificial neural network and genetic algorithm

17-4PH alloy is a precipitation hardening stainless steel with a desirable combination of high corrosion resistance and good mechanical properties at temperatures up to 300 °C in engineering applications. Ageing mechanism in this alloy is rather complicated, hence, for optimum hardness, selection of heat treatment parameters is critical. In this investigation, first, Artificial Neural Network (ANN) was used to model the relationship between ageing times and temperatures with the corresponding hardness. R2 value was 0.9747 in the final ANN model indicating this model could predict hardness values, appropriately. Then, Genetic Algorithm (GA) was used to find optimum aging time and temperature for the target of maximum hardness value. ANN model was used as the fitness function in GA. According to the GA-ANN simulation, a hardness of 44.9 HRC would be achieved by ageing 129 min at 464 °C. Finally, the model was validated by conducting heat treatment experiments carried out using the predicted parameters. A maximum hardness of 44.1 HRC was obtained in the experimental work, showing a difference of 1.8% with the proposed model.