Scholarlink

<p>Electrical Discharge Machining (EDM) is a groundbreaking process that excels in producing high-precision components from challenging materials. Traditionally, optimizing process parameters has been costly and time-consuming. However, Acoustic Emission (AE) technology offers a powerful solution for real-time monitoring, revealing critical insights into electrode material removal. This study explores the relationship between AE signals and erosion mechanisms in EDM using frequency band analysis and spectral entropy. Conducted on AISI H13 steel workpieces, the experiments varied current and pulse duration using electrolytic copper and graphite electrodes with either hydrocarbon oil or deionized water as dielectrics. The excitation band results successfully differentiated and isolated the frequency ranges directly associated with the loss of each material (copper at 50–90&#xa0;kHz, steel at 135–185&#xa0;kHz, and graphite at 215–250&#xa0;kHz). The low standard deviation of the mean spectral entropy for each material—maximum deviation of 2.5% for copper and 1.7% for graphite—demonstrates the repeatability of the entropy measurements for EDM monitoring. Lower spectral entropy values correlated with higher material removal rates and electrode wear, confirming the reliability of the method. The combined use of AE monitoring and spectral entropy analysis serves as an effective indicator for evaluating EDM process efficiency, enabling quantitative characterization of discharge behavior and material removal performance under varying machining conditions. To overcome empirical parameter tuning, a data-driven diagnostic framework is proposed, scalable and compatible with industrial monitoring systems. This approach enables real-time and predictive EDM analysis, supporting integration into machine control and improving process efficiency and precision.</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

Spectral entropy of acoustic emission signals for monitoring electric discharge machining performance

  • 期刊论文

  • 发表日期：

    2026年1月5日

查看全文
  • Samuel Soares Ferreira,
  • Luís Henrique Andrade Maia,
  • Fred Lacerda Amorim

摘要

Electrical Discharge Machining (EDM) is a groundbreaking process that excels in producing high-precision components from challenging materials. Traditionally, optimizing process parameters has been costly and time-consuming. However, Acoustic Emission (AE) technology offers a powerful solution for real-time monitoring, revealing critical insights into electrode material removal. This study explores the relationship between AE signals and erosion mechanisms in EDM using frequency band analysis and spectral entropy. Conducted on AISI H13 steel workpieces, the experiments varied current and pulse duration using electrolytic copper and graphite electrodes with either hydrocarbon oil or deionized water as dielectrics. The excitation band results successfully differentiated and isolated the frequency ranges directly associated with the loss of each material (copper at 50–90 kHz, steel at 135–185 kHz, and graphite at 215–250 kHz). The low standard deviation of the mean spectral entropy for each material—maximum deviation of 2.5% for copper and 1.7% for graphite—demonstrates the repeatability of the entropy measurements for EDM monitoring. Lower spectral entropy values correlated with higher material removal rates and electrode wear, confirming the reliability of the method. The combined use of AE monitoring and spectral entropy analysis serves as an effective indicator for evaluating EDM process efficiency, enabling quantitative characterization of discharge behavior and material removal performance under varying machining conditions. To overcome empirical parameter tuning, a data-driven diagnostic framework is proposed, scalable and compatible with industrial monitoring systems. This approach enables real-time and predictive EDM analysis, supporting integration into machine control and improving process efficiency and precision.