Scholarlink

This study presents a hybrid optimization approach combining the Non-dominated Sorting Genetic Algorithm II (NSGA-II) and the Entropy-Assisted Additive Ratio Assessment (EAMR) method for enhancing the performance of ultrasonic vibration-assisted electrical discharge machining (UV-EDM) on 90CrSi tool steel. The process employs copper electrodes with the dual objective of minimizing surface roughness (Ra) and maximizing material removal rate (MRR). A Box–Behnken design (BBD) was used to develop predictive models for process responses. NSGA-II was implemented to generate a Pareto front representing the trade-off between Ra and MRR, while the EAMR method was applied to identify the best compromise solution using entropy-based weighting of criteria. The optimal parameters identified are: vibration amplitude (A) = 0.800 µm, pulse-on time (Ton) = 12.365 µs, pulse-off time (Toff) = 11.135 µs, peak current (IP) = 5.294 A, and spark voltage (SV) = 5.023 V. Under these conditions, the achieved material removal rate is 4.134 mm3/min and the corresponding surface roughness is 4.302 µm. The proposed approach effectively balances machining efficiency and surface quality, offering practical guidance for the optimization of UV-EDM processes for hard tool steels.

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

Multi-objective Optimization of Ultrasonic Vibration-Assisted EDM on 90CrSi Tool Steel Using NSGA-II and EAMR

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Dinh Van Thanh,
  • Le Thu Quy,
  • Vu Ngoc Pi,
  • Tran Thi Phuong Thao

摘要

This study presents a hybrid optimization approach combining the Non-dominated Sorting Genetic Algorithm II (NSGA-II) and the Entropy-Assisted Additive Ratio Assessment (EAMR) method for enhancing the performance of ultrasonic vibration-assisted electrical discharge machining (UV-EDM) on 90CrSi tool steel. The process employs copper electrodes with the dual objective of minimizing surface roughness (Ra) and maximizing material removal rate (MRR). A Box–Behnken design (BBD) was used to develop predictive models for process responses. NSGA-II was implemented to generate a Pareto front representing the trade-off between Ra and MRR, while the EAMR method was applied to identify the best compromise solution using entropy-based weighting of criteria. The optimal parameters identified are: vibration amplitude (A) = 0.800 µm, pulse-on time (Ton) = 12.365 µs, pulse-off time (Toff) = 11.135 µs, peak current (IP) = 5.294 A, and spark voltage (SV) = 5.023 V. Under these conditions, the achieved material removal rate is 4.134 mm3/min and the corresponding surface roughness is 4.302 µm. The proposed approach effectively balances machining efficiency and surface quality, offering practical guidance for the optimization of UV-EDM processes for hard tool steels.