Scholarlink

The main objective of this contribution is to identify, implement, and compare deep learning methods for the recognition of acoustic emission signals. Introduction of these modern methods can improve performance in industrial utilisations of AE, such as defectoscopy, monitoring, or machining process control. An experiment was conducted to obtain relevant data to compare selected neural network architectures. Five architectures designed to work directly with 1D signals as an input data are presented. These models are compared based on their performance in a classification task using the data from the experiment. Additionally, an adapted version of the best-performing Pooled Inception Time network is utilized in a regression task to predict continuous dependent variable.

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

Modern Deep Neural Networks in Acoustic Emission Signal Analysis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jan Zavadil,
  • Václav Kůs,
  • Milan Chlada

摘要

The main objective of this contribution is to identify, implement, and compare deep learning methods for the recognition of acoustic emission signals. Introduction of these modern methods can improve performance in industrial utilisations of AE, such as defectoscopy, monitoring, or machining process control. An experiment was conducted to obtain relevant data to compare selected neural network architectures. Five architectures designed to work directly with 1D signals as an input data are presented. These models are compared based on their performance in a classification task using the data from the experiment. Additionally, an adapted version of the best-performing Pooled Inception Time network is utilized in a regression task to predict continuous dependent variable.