Scholarlink

<p>The focus of this paper investigates the Wire Electrical Discharge Machining (WEDM) performance of Al-6063/ Fly Ash / ZrO₂ hybrid composite specimens on Material Removal Rate (MRR) and Surface Roughness (SR). The experiments were performed using the three parameters current (60, 70, 80&#xa0;A), wire feed rate (4,6 and 8&#xa0;mm/min) and pulse on time (50,75 and 100 µs). ANFIS was employed to minimize predictive error and maximize process optimization by simulating and predicting machining responses. The ANFIS learning curve showed good signs of learning, with a rapid drop of the MRR training error from 9.7 × 10⁻⁶ to about 9.6 × 10⁻⁶ in the first 20 epochs. The experimental results revealed that the highest MRR was attained at high peak current, medium wire feed rate, and medium-high pulse-on time, which supported the combined effect of heat and discharging actions on material removal. On the other hand, an increased current and a longer pulse-on time led to higher surface roughness. The high correlation between actual and expected results, indicated with R² values of 0.95 to 0.98, proves that ANFIS model is a robust tool for the accurate prediction of WEDM performance characteristics. The contribution of this research is a complete guideline for the betterment of Hybrid composites using aluminium in precision machining.</p>

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

Advanced predictive analytics for machining performance of Al-6063 hybrid composites reinforcing fly Ash and ZrO₂ using ANFIS simulation

  • 期刊论文

  • 发表日期：

    2026年2月9日

查看全文
  • Anil Kumar Deepati,
  • S. Ravichandran,
  • Chandra Shekhar Verma,
  • Ravi Ranjan Kumar,
  • Sujeet Kumar

摘要

The focus of this paper investigates the Wire Electrical Discharge Machining (WEDM) performance of Al-6063/ Fly Ash / ZrO₂ hybrid composite specimens on Material Removal Rate (MRR) and Surface Roughness (SR). The experiments were performed using the three parameters current (60, 70, 80 A), wire feed rate (4,6 and 8 mm/min) and pulse on time (50,75 and 100 µs). ANFIS was employed to minimize predictive error and maximize process optimization by simulating and predicting machining responses. The ANFIS learning curve showed good signs of learning, with a rapid drop of the MRR training error from 9.7 × 10⁻⁶ to about 9.6 × 10⁻⁶ in the first 20 epochs. The experimental results revealed that the highest MRR was attained at high peak current, medium wire feed rate, and medium-high pulse-on time, which supported the combined effect of heat and discharging actions on material removal. On the other hand, an increased current and a longer pulse-on time led to higher surface roughness. The high correlation between actual and expected results, indicated with R² values of 0.95 to 0.98, proves that ANFIS model is a robust tool for the accurate prediction of WEDM performance characteristics. The contribution of this research is a complete guideline for the betterment of Hybrid composites using aluminium in precision machining.