Scholarlink

<p>The short action time (10–4 ~ 10–2&#xa0;s) and small material processing layer (10–4 ~ 10–2&#xa0;m) of the laser polishing process pose challenges for online monitoring. To address this, this paper first proposes using acoustic emission technology for online monitoring of the laser polishing process. It analyzes the influence of different process parameters on the acoustic emission signal law and uses acoustic emission signal parameters to characterize the laser polishing process in different processing states. This enables online monitoring and intelligent regulation/control of the laser polishing process, providing theoretical support. The acoustic emission signals were analyzed in the time–frequency domain to identify differences in frequency domain distribution between the polished and noise signals, and noise reduction was performed using the wavelet hard thresholding algorithm. The correspondence between the laser processing state and the time–frequency characteristics of the acoustic emission signal was analyzed, and the characteristic parameters with significant variations were identified as the input variables of the processing state classification and identification model based on the BP neural network algorithm. The results show that the SSM and SOM states of the polishing process can be accurately identified with an accuracy rate of 97.7%.</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

Acoustic Emission Based Classification of Processing States for Laser Polishing

  • 期刊论文

  • 发表日期：

    2025年2月18日

查看全文
  • Fanming Guo,
  • Yanhou Liu,
  • Pengwei Wang,
  • Juan Ma,
  • Guiguan Zhang,
  • Jinguo Han,
  • Dehui Wang,
  • Longxing Yin

摘要

The short action time (10–4 ~ 10–2 s) and small material processing layer (10–4 ~ 10–2 m) of the laser polishing process pose challenges for online monitoring. To address this, this paper first proposes using acoustic emission technology for online monitoring of the laser polishing process. It analyzes the influence of different process parameters on the acoustic emission signal law and uses acoustic emission signal parameters to characterize the laser polishing process in different processing states. This enables online monitoring and intelligent regulation/control of the laser polishing process, providing theoretical support. The acoustic emission signals were analyzed in the time–frequency domain to identify differences in frequency domain distribution between the polished and noise signals, and noise reduction was performed using the wavelet hard thresholding algorithm. The correspondence between the laser processing state and the time–frequency characteristics of the acoustic emission signal was analyzed, and the characteristic parameters with significant variations were identified as the input variables of the processing state classification and identification model based on the BP neural network algorithm. The results show that the SSM and SOM states of the polishing process can be accurately identified with an accuracy rate of 97.7%.