基于改进SSA-VMD的列车门丝杠机构故障诊断技术研究

Research on Fault Diagnosis Technology of Train Door Screw Mechanism Based on Improved SSA-VMD

【目的】为解决地铁运行时因列车门噪声过大而导致故障诊断难的问题,以列车门为研究对象,提出一种基于改进麻雀搜索(SSA)算法的变分模态分解(VMD)振动信号降噪法,并通过支持向量机来对故障进行诊断。【方法】首先,利用Hénon混沌映射来初始化种群,将非线性权重因子引入群体行为阶段,并通过Levy飞行策略及柯西变异对位置进行更新。其次,通过改进的麻雀搜索算法对变分模态分解算法中的惩罚因子α和模态分解数K进行全局寻优,确定参数分解并重构,得到降噪信号。最后,使用主成分分析法(PCA)来提取特征,并利用支持向量机(SVM)来诊断故障。【结果】试验结果表明,该方法对振动信号的降噪效果明显,故障诊断准确率达91%,验证了该方法的有效性。【结论】该方法能有效克服传统VMD去噪参数难以选取的问题,对列车门故障诊断研究具有一定的参考价值。

[Purposes]In order to solve the problem of difficult fault diagnosis caused by excessive noise of train doors during subway operation,a variational mode decomposition(VMD)vibration signal denoising method based on improved sparrow search(SSA)algorithm is proposed,and the fault is diagnosed by support vector machine.[Methods]Firstly,the Hénon chaotic map is used to initialize the population,the nonlinear weight factor is introduced into the group behavior stage,and the position is updated by Levy flight strategy and Cauchy mutation.Secondly,through the improved sparrow search algorithm,the penalty factorαand the modal decomposition number K in the variational modal decomposition algorithm are globally optimized,the parameter decomposition and reconstruction are determined,and the noise reduction signal is obtained.Finally,principal component analysis(PCA)is used to extract features,and support vector machine(SVM)is used to diagnose faults.[Findings]The experimental results show that the method has obvious noise reduction effect on vibration signals,and the accuracy of fault diagnosis is 91%,which verifies the effectiveness of the method.[Conclusions]This fectively overcome the problem that the traditional VMD denoising parameters are difficult to select,which has certain reference value for the research of train door fault diagnosis.