基于小波时频图和卷积神经网络的断路器故障诊断分析

高压断路器操动机构振动信号为非平稳性信号,蕴含着丰富的操动机构工作状态的信息,对操动机构工作状态的检验辨识具有重大意义。提出一种基于小波时频图和卷积神经网络的断路器故障诊断方法。对操动机构振动信号进行连续小波变换生成时频图(CWT),并对时频图进行统一压缩预处理;将预处理后的时频图作为特征图输入卷积神经网络AlexNet模型;通过对网络参数的调整,逐步改进网络模型,有监督地实现对操动机构故障状态的辨识诊断。结果表明,该方法能够有效地运用于断路器操动机构故障辨识诊断,与小波频带能量-RBF、小波频带能量-SVM的故障识别相比,故障识别准确率最高。

Fault diagnosis and analysis of circuit breaker based on wavelet time-frequency representations and convolution neural network

The vibration signal of the operating mechanism of high voltage circuit breaker is non-stationary, which contains abundant information of the operating state of the operating mechanism.It is of great significance for the inspection and identification of the operating state of the operating mechanism.This paper presented a new method of applying wavelet time-frequency diagram and convolution neural network to fault diagnosis of circuit breakers.Continuous wavelet transform(CWT) was used to analyze vibration signals of operating mechanism and get time-frequency representation.And the time-frequency representations were compressed to the appropriate size.After that, all the compressed time-frequency representations were taken as input feature maps of convolution neural network AlexNet model.By adjusting the network parameters, the network model was improved step by step, and the fault state identification and diagnosis of the operating mechanism was realized with supervision.The results show that this method can be effectively applied to fault identification and diagnosis of circuit breaker operating mechanism.Compared with wavelet band energy-RBF, wavelet band energy-SVM algorithm, the accuracy of the method in the paper is the highest.