Scholarlink

<p>This study investigates the condition monitoring of milling tools using vibration analysis for detecting wear and damage. Milling tools experience progressive deterioration during machining, affecting surface quality and dimensional accuracy. Multiaxial vibration signals were collected from four tools representing different wear conditions: New Bit, Mild Wear, Medium Wear, and Heavy Wear. The data were analyzed using power spectral density (PSD) in the frequency domain and continuous wavelet transform (CWT) in the time–frequency domain. PSD analysis revealed that frequency peaks above 4000&#xa0;Hz, particularly along the transverse (y-axis) direction, were closely associated with tool wear. For Mild and Medium Wear, dominant peaks appeared at 4941 and 4990&#xa0;Hz, respectively, indicating that increased tool degradation corresponds to higher vibration frequencies. However, the Heavy Wear tool generated weak vibrations due to inefficient material removal. In contrast, CWT exhibited superior capability in identifying sudden tool fractures through sharp increases in wavelet energy, reaching 9.24 × 10<sup>3</sup>&#xa0;(m/s<sup>2</sup>)<sup>2</sup>, caused by amplitude spikes. These findings demonstrate that wavelet transform offers enhanced sensitivity and diagnostic accuracy compared to conventional frequency domain analysis, making it a reliable approach for real-time condition monitoring and predictive maintenance of milling tools in industrial machining environments.</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

Milling Tool Health Monitoring Via Power Spectral and Wavelet Energy Analysis of Multiaxial Vibrations

  • 期刊论文

  • 发表日期：

    2026年3月6日

查看全文
  • C. H. Chin,
  • S. Abdullah

摘要

This study investigates the condition monitoring of milling tools using vibration analysis for detecting wear and damage. Milling tools experience progressive deterioration during machining, affecting surface quality and dimensional accuracy. Multiaxial vibration signals were collected from four tools representing different wear conditions: New Bit, Mild Wear, Medium Wear, and Heavy Wear. The data were analyzed using power spectral density (PSD) in the frequency domain and continuous wavelet transform (CWT) in the time–frequency domain. PSD analysis revealed that frequency peaks above 4000 Hz, particularly along the transverse (y-axis) direction, were closely associated with tool wear. For Mild and Medium Wear, dominant peaks appeared at 4941 and 4990 Hz, respectively, indicating that increased tool degradation corresponds to higher vibration frequencies. However, the Heavy Wear tool generated weak vibrations due to inefficient material removal. In contrast, CWT exhibited superior capability in identifying sudden tool fractures through sharp increases in wavelet energy, reaching 9.24 × 103 (m/s2)2, caused by amplitude spikes. These findings demonstrate that wavelet transform offers enhanced sensitivity and diagnostic accuracy compared to conventional frequency domain analysis, making it a reliable approach for real-time condition monitoring and predictive maintenance of milling tools in industrial machining environments.