【摘要】 提出了一种基于小波分析和BP神经网络的滚动轴承故障诊断方法,首先采用小波包对滚动轴承振动信号进行分解与重构,然后提取重构后振动信号的峭度值,将峭度值作为特征参数输入神经网络,进行故障模式识别。通过对实验数据的分析信号表明,能有效地识别滚动轴承工作状态与故障类型。更多还原

【Abstract】 A fault-diagnosis method for roller bearings based on wavelet analysis and BP neural network is proposed. Firstly, roller bearing signal is decomposed and reconstructed with wavelet packets;then, the Kurtosis factors of the reconstructed signals are extracted and served as characteristic parameters to be put into neural network;finally, the work condition and fault pattern are identified by the output of the neural network. Analysis returns of the experimental data show that, the proposed method in this paper can be applied in roller bearing fault diagnosis efficiently.更多还原