基于模糊神经网络的铁路车辆滚动轴承故障诊断研究

Research on fault diagnosis of rolling bearings in railway vehicles based on fuzzy neural network

滚动轴承是铁路车辆运行中关键的组件之一。然而,在使用过程中,滚动轴承可能会受到疲劳、摩擦、腐蚀等多种故障和损伤的影响。如果这些问题不能及时检测和诊断,将可能导致运输中断,甚至造成严重事故和经济损失。为此,文章提出了1种基于模糊神经网络的故障诊断方法,用于铁路车辆滚动轴承的故障监测与判断。该方法首先应用小波阈值去噪算法来抑制采集到的滚动轴承噪声数据,从而提高信号质量。接着通过特征提取算法分析去噪结果,以获取滚动轴承故障的关键特征。最后利用模糊神经网络建立故障诊断模型,并将提取到的特征输入模型进行学习训练,从而得到精确的滚动轴承故障诊断结果。通过实验验证,该方法在滚动轴承故障诊断方面表现出良好的准确性和可靠性,具有广泛的应用前景。

A fault diagnosis method based on fuzzy neural network is proposed for the fault monitoring and judgment of rolling bearings in railway vehicles.This method first applies wavelet threshold denoising algorithm to suppress the collected rolling bearing noise data,thereby improving signal quality.Next,the denoising results are analyzed using feature extraction algorithms to obtain the key features of rolling bearing faults.Finally,a fault diagnosis model is established using a fuzzy neural network,and the extracted features are input into the model for learning and training,in order to obtain accurate rolling bearing fault diagnosis results.Through experimental verification,this method has shown good accuracy and performance in the diagnosis of rolling bearing faults,and has broad application prospects.