Scholarlink

This study investigates the modeling and constrained optimization of electrode wear rate (EWR) in powder-mixed electrical discharge machining (PMEDM) of hardened SKD11 tool steel using SiC powder. A series of experiments were conducted by varying key process parameters, including powder concentration (Cp), particle size (Sp), pulse-on time (Ton), pulse-off time (Toff), peak current (Ip), and servo voltage (SV). A stepwise regression method was employed to construct a reduced pure quadratic response surface model (RSM) for EWR, allowing for statistical validation and interpretability. The final model achieved a high coefficient of determination (R2 = 1.0000, adjusted R2 = 0.9999), with most terms showing statistical significance. Residual plots confirmed the model’s adequacy for prediction. An optimization procedure using fmincon was applied to minimize EWR under the practical constraint EWR ≥ 1 mg/min. The optimal solution was found at Cp = 0.0540 g/l, Sp = 243.74 µm, Ton = 29.93 µs, Toff = 18.98 µs, Ip = 8.76 A, and SV = 4.19 V. The proposed approach provides a reliable framework for predicting and minimizing tool wear in PMEDM, contributing to enhanced tool life and process sustainability.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction and Minimization of Electrode Wear in Powder-Mixed EDM Using Stepwise RSM Approach

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Dinh Van Thanh,
  • Nguyen Manh Cuong,
  • Do Thi Tam,
  • Hoang Xuan Tu

摘要

This study investigates the modeling and constrained optimization of electrode wear rate (EWR) in powder-mixed electrical discharge machining (PMEDM) of hardened SKD11 tool steel using SiC powder. A series of experiments were conducted by varying key process parameters, including powder concentration (Cp), particle size (Sp), pulse-on time (Ton), pulse-off time (Toff), peak current (Ip), and servo voltage (SV). A stepwise regression method was employed to construct a reduced pure quadratic response surface model (RSM) for EWR, allowing for statistical validation and interpretability. The final model achieved a high coefficient of determination (R2 = 1.0000, adjusted R2 = 0.9999), with most terms showing statistical significance. Residual plots confirmed the model’s adequacy for prediction. An optimization procedure using fmincon was applied to minimize EWR under the practical constraint EWR ≥ 1 mg/min. The optimal solution was found at Cp = 0.0540 g/l, Sp = 243.74 µm, Ton = 29.93 µs, Toff = 18.98 µs, Ip = 8.76 A, and SV = 4.19 V. The proposed approach provides a reliable framework for predicting and minimizing tool wear in PMEDM, contributing to enhanced tool life and process sustainability.