基于Mallat小波分解与改进GWO-SVM的道岔故障诊断

Turnout fault diagnosis based on Mallat wavelet decomposition and improved GWO-SVM

为顺应道岔故障诊断向智能化和自动化的发展趋势,以S700K转辙机功率曲线为例,提出一种改进型灰狼优化算法(GWO)与支持向量机(SVM)相结合的故障诊断方法。在8种故障模式和正常模式所对应的功率曲线实施5层Mallat小波分解,得到各层近似系数和细节系数,并计算各层系数的平方和;利用主成分分析法对系数平方和组成的向量集进行降维,得到特征向量集;改进型灰狼优化算法优化支持向量机参数,并用优化好的支持向量机进行故障预测。研究结果表明:该方法能有效提高道岔故障诊断的准确率。

In order to adapt to the development trend of intelligent fault diagnosis, this paper took the power curve of S700 K switch as an example, and proposed a fault diagnosis method based on improved Gray Wolf Optimization Algorithm(GWO) and Support Vector Machine(SVM). Firstly, five layers of Mallat wavelet decomposition were performed on the power curves corresponding to the eight failure modes and the normal mode, and the approximate coefficients and detail coefficients of each layer were obtained. The square sum of the coefficients of each layer was calculated. Secondly, the principal component analysis method was used to reduce the dimension of the vector set composed of the squares of the coefficients to obtain the feature vector set. Finally, the improved gray wolf optimization algorithm optimized the support vector machine parameters, and the optimized support vector machine was used to predict the fault. The simulation results show that the method can effectively improve the accuracy of turnout fault diagnosis.