Scholarlink

<p>This study presents an intelligent system for diagnosing the severity of parallel misalignment in worm gear reducers using vibration analysis. Accelerometer data provided features such as frequency amplitude, harmonics, peaks, and power spectral density to train a multi-layer perceptron neural network with backpropagation. The system classifies conditions as “reference,” “intermediate,” or “serious,” achieving 98.8% accuracy. Results confirm the system’s potential as a support tool for predictive maintenance in low-power gear-driven machines.</p> Graphical Abstract <p>Parallel misalignments identification scheme. </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

Intelligent system for identification of parallel misalignments in gears of the worm type crown using ANN

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月5日

查看全文
  • João Manoel de Oliveira Neto,
  • Marcelo Cavalcanti Rodrigues,
  • Antônio Almeida Silva,
  • Andersson Guimarães Oliveira

摘要

This study presents an intelligent system for diagnosing the severity of parallel misalignment in worm gear reducers using vibration analysis. Accelerometer data provided features such as frequency amplitude, harmonics, peaks, and power spectral density to train a multi-layer perceptron neural network with backpropagation. The system classifies conditions as “reference,” “intermediate,” or “serious,” achieving 98.8% accuracy. Results confirm the system’s potential as a support tool for predictive maintenance in low-power gear-driven machines.

Graphical Abstract

Parallel misalignments identification scheme.