Ensemble Machine learning based performance prediction of electrical discharge machining for AISI D6 tool steel

Abstract

Die and mould makers frequently employ AISI-D6 steel. First, 32 trials of electrical discharge machining (EDM) are carried out on the aforementioned material. A variety of EDM procedure performance measures, including tool wear ratio (TWR), material removal rate (MRR), and surface roughness (Ra), are taken into account, as well as the pulse time (Ton), pulse current (I), and pulse voltage (V). Operators are guided by process. Increasing MRR and Ra concentrations and decreasing TWR concentrations were found to increase pulse on-time values. As a result of increasing the pulse current, physical removal rate, tool wear ratio and surface unevenness increased. As a result of the increased voltage, the MRR, TWR, and Ra values were all reduced. The outcomes are predicted using ensemble machine learning models. The experimental data used to create estimation models for MRR, TWR, and Ra. In addition, an unknown set of experiments has validated the model-based predictions.