Modeling surface quality, cost and energy consumption during milling of alloy 2017A: a comparative study integrating GA-ANN and RSM models

Abstract

ABSTRACT The use of simple machining features to determine cutting parameters and the machining process is limited, as parts may contain complex features interacting with each other. This study, therefore, focuses on pocket/groove features and proposes an approach integrating hybrid GA-ANN and RSM algorithms to predict surface quality, cost, and energy consumption (QCE). A parametric study was carried out, taking into account the swarm population size (pop) and the number of neurons (n) in the hidden layer, to find the best prediction using the hybrid GA-ANN algorithm. The results showed the highest correlation values (R2) for all output variables (above 0.97%). The study also revealed that the allocation of machining strategies and sequences can have a significant impact on energy consumption, with a 99.25% variation between minimum and maximum values. Mean square error (MSE) data confirmed the effectiveness of the GA-ANN model. Compared with RSM model predictions, energy consumption (Etot), cost (Ctot), and surface quality (Ra) values all showed statistically significant increases of 90.9%, 96.55%, and 40.18%, respectively. This study highlights the potential of the GA-ANN hybrid approach for multi-criteria prediction (quality, cost, and energy: QCE) in comparison with the RSM method, offering potential improvements for machining 2017A alloy.