Improved PSO-SVM-Based Fault Diagnosis Algorithm for Wind Power Converter

Due to the complexity of the working environment of wind power systems, wind power converters (WPCs) will experience different types of failures. In order to improve the reliability and stability of the wind power system and avoid system shutdown, this paper takes the back-to-back wind power converter as the research object, and proposes an improved particle swarm optimization (PSO)- support vector machine (SVM) algorithm, which solves the problems of low accuracy, long execution time and high data dependence of traditional fault diagnosis schemes. Firstly, the moving average algorithm is used to process the fault data of the converter, and then the support vector machine based on the improved particle swarm algorithm is used to diagnose the open-circuit fault of the converter. In the dataset consisting of 9800 training samples and 4200 testing samples, the accuracy of training samples is 98.898%, and the accuracy of testing samples is 98.4524%, which effectively improves the problem that traditional SVM can only handle small batch nonlinear datasets. Finally, the effectiveness and superiority of the proposed scheme are verified by comparing the SVM algorithm under the same dataset.