Research on Electromechanical Equipment Fault Prediction

摘要：

This paper, taking DC motor as a typical example of electromechanical equipment, analyzes malfunction law of equipment and researches on its structural feature and the symptom parameters of faults. The fault prediction models have been established that based on wavelet analysis and artificial neural network. By the advantages of Wavelet analyzing the signal in different resolutions, wavelet analysis is used to extract the symptoms of potential faults. Artificial Neural Network (ANN) has a good performance to process the non-linear mapping problem, therefore historical data and symptom parameters are used to train BP network and the symptom parameter prediction model and symptom parameter - fault mapping model can be established. In the end, analysis and comparison on predicting and experimental findings has justified that the wavelet-ANN model can attain the functions of extracting symptoms, predicting symptom parameters and fault mapping.

关键词： electromechanical equipment fault prediction wavelet analysis BP network

作者: LI Hongru YE Peng HUANG Kun

作者单位: Dept. of Missile Engineering, Ordnance Engineering College, Shijiazhuang,050003, China

会议类型: 国际会议

会议名称: 第七届国际测试技术研讨会

会议地点: 北京

会议语种:英文

在线出版日期: 2007-08-05（万方平台首次上网日期，不代表论文的发表时间）