Sustainability metrics targeted optimization and electric discharge process modelling by neural networks

Aluminium and its alloys, especially Al6061, have gathered significant interest among researchers due to its less density, great durability, and high strength. Due to their lightweight properties, the precise machining of these alloys can become expensive through conventional machining operations for intricate products. Therefore, non-traditional machining such as electric discharge machining (EDM) can potentially be opted for the cutting of Al6061. EDM is often criticized due to its low machining rates, therefore, in the current work, cryogenic treatment (CT) has been performed on the brass electrode to evaluate the improvement in the machining rates. In addition, kerosene oil (KO) has been engaged in traditional EDM which is replaced with the deionized water (DI) based dielectric as a sustainable alternative. The machining variables such as spark voltage (SV), pulse-on-time (PON), peak current (IP), and Al2O3 powder concentration (CP) have been chosen to determine the material removal rate (MRR), surface roughness (SR), and specific energy consumption (SEC) while comparing non-treated (NT), and cryogenically treated (CT) brass electrodes during EDM. The results were analyzed through optical micrographs, scanning electron microscopy (SEM) analysis, energy dispersive x-ray (EDX) examination, and 3D surface plots. An artificial neural network (ANN) has been constructed for the better prediction of output responses. Moreover, multi-response optimization through the non-dominated sorting genetic algorithm (NSGA-II) has also been performed. The magnitudes of MRRCT, SRCT, and SECCT obtained by multi-response optimization are 64.82%, 27.45%, and 46.60% are better than the values obtained by un-optimized settings of CT brass electrodes. However, the optimal magnitudes of processing parameters are IP = 24.85 A, SV = 2.18 V, PON = 119.11 mu s, and CP = 1.05 g/100 ml.