| 基于图卷积网络的电机故障诊断研究 |
投稿时间：2024-03-20 修订日期：2024-05-28  点此下载全文 |
| 引用本文： |
| 摘要点击次数: 37 |
| 全文下载次数: 0 |
|
|
| 中文摘要:振动信号常被用来分析和判断电机在运行中出现的故障类型。当前故障数据的处理方法主要是通过专家判断分析，处理成本较高。近年来，基于深度学习的电机故障数据处理方法已开始用于电机的故障诊断任务。本文提出了一种多头注意力增强的图卷积神经网络深度学习模型，对输入的故障节点数据进行编码，通过递归神经网络和非线性层实现电机的故障类型分类，利用时序故障数据判断电机的故障类型。实验结果表明，提出的模型在电故障诊断任务中取得了良好效果。 |
| 中文关键词:振动信号 电机 故障诊断 图卷积神经网络 |
| |
| Research on Motor Fault Diagnosis Based on Graph Convolutional Network |
|
|
| Abstract:Vibration signals are often used to analyze and judge the types of faults that occur in the operation of induction motors. At present, the processing method of fault data is mainly through expert judgment and analysis, and the processing cost is high. In recent years, deep learning-based induction motor fault data processing methods have been used in induction motor fault diagnosis tasks. In this paper, a deep learning model of graph convolutional neural network based on multi-head attention enhancement is proposed to determine the fault type of induction motor by using time series fault data. The proposed method firstly obtains the extended time sequence feature representation, encodes the input fault node data based on graph convolutional network, and finally realizes the fault type classification of induction motor through recurrent neural network and nonlinear layer. Experimental results show that the proposed model can achieves the better performance. |
| keywords:Vibration signals Motors Fault diagnosis Graph convolutional Network |
| 查看全文 查看/发表评论 下载pdf阅读器 |
