摘要
根据检测设备采集的数传设备故障样本信号,采用小波分析方法,提取信号的小波能量熵,并将其与其他特征参数一起形成特征向量,利用训练好的BP神经网络模型对设备故障进行诊断,从而确定设备故障模块。结果表明,该方法在数传设备故障诊断中具有较高的故障诊断率。
According to the number of fault sample signal of digital Transmission equipment,using the method of wavelet analysis,the wavelet energy entropy of signal is extracted in the article. The entropy itself and other characteristic parameters are formed together to be a feature vector. The trained BP neural network model is used to diagnose the equipment fault so as to determine the fault module. The results show that this method has a high rate of fault diagnosis in the fault diagnosis for digital transmission equipment.
出处
《山西电子技术》
2016年第3期90-92,共3页
Shanxi Electronic Technology
基金
国家自然科学基金资助项目(60874112)
关键词
小波能量熵
BP神经网络
故障诊断
wavelet
energy entropy
BP neural network
fault diagnosis
