Scholarlink

<p>The increasing demand for dies and molds in the manufacturing sector requires a smart solution to enhance productivity while ensuring responsible energy consumption and production material. This study focuses on improving the electric discharge machining (EDM) process, which is essential in the production of tooling. Among tool steel materials, D2 is considered in the work because of its wide applications in the production tooling sector. EDM is an electro-thermal erosion process that uses a dielectric as a conducting medium in the erosion process. In this work, novel dielectrics are evaluated (deionized water and surfactants (Tween-80, Span-80)) against their dimensional accuracy and energy consumption requirements with pulse current, water concentration (W/C), and surfactant concentration (S/C) as control parameters under statistical design of experiments. EDM’s complex physical phenomenon is modelled using machine learning’s artificial neural networks. The process is optimized using Multi-Objective Genetic Algorithm-Based Optimization (MOGA), and validated with confirmatory experiments and microscopic analyses. The application of novel dielectric demonstrated a 6.52% reduction in specific energy consumption and an 18.75% increase in dimensional accuracy at water concentration = 9.54%/L, surfactant concentration = 0.06&#xa0;g/L, and pulse current = 11.9125 A.</p> Graphical Abstract <p></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

Investigating novel dielectrics in electric discharge machining of D2 tool steel using experimental and machine learning approaches

  • 期刊论文

  • 发表日期：

    2026年1月10日

查看全文
  • Haider Ali Bhatti,
  • Tayyiba Rashid,
  • Anamta Khan

摘要

The increasing demand for dies and molds in the manufacturing sector requires a smart solution to enhance productivity while ensuring responsible energy consumption and production material. This study focuses on improving the electric discharge machining (EDM) process, which is essential in the production of tooling. Among tool steel materials, D2 is considered in the work because of its wide applications in the production tooling sector. EDM is an electro-thermal erosion process that uses a dielectric as a conducting medium in the erosion process. In this work, novel dielectrics are evaluated (deionized water and surfactants (Tween-80, Span-80)) against their dimensional accuracy and energy consumption requirements with pulse current, water concentration (W/C), and surfactant concentration (S/C) as control parameters under statistical design of experiments. EDM’s complex physical phenomenon is modelled using machine learning’s artificial neural networks. The process is optimized using Multi-Objective Genetic Algorithm-Based Optimization (MOGA), and validated with confirmatory experiments and microscopic analyses. The application of novel dielectric demonstrated a 6.52% reduction in specific energy consumption and an 18.75% increase in dimensional accuracy at water concentration = 9.54%/L, surfactant concentration = 0.06 g/L, and pulse current = 11.9125 A.

Graphical Abstract