Fault diagnosis of WOA-SVM high voltage circuit breaker based on PCA Principal Component Analysis

Abstract

High voltage circuit breakers (HVCB) are significant protection and control devices for electric systems, and its operating state directly influences the stability and reliability of electric systems. In order to solve the problem of low precision of the mechanical diagnostic accuracy of HVCB, this paper propose a means of optimization based on the principal component analysis (PCA), and the diagnosis of the HVCB is supplemented by the support vector machine (SVM) of the whale algorithm (WOA) optimization. Firstly, the present machine fault state of the circuit breaker is experimentally simulated, the LabVIEW software is used to obtain the vibration signal of different states, then the Hilbert Huang conversion is made to the collected vibration signal to obtain an appropriate envelope spectrum, and the entropy of the envelope spectrum is calculated. Then, we use the PCA method to calculate the contribution rate of each characteristic value,and the feature values with a cumulative contribution rate higher than 90% are used as the final feature vector . Finally , use the WOA algorithm to optimize the penalty factors and kernel parameters of SVM. Through parameter optimization and feature optimization,it can be found that the WOA-SVM was superior to GA-SVM for optimized time and accuracy, providing an effective method for fault diagnosis of the circuit breaker.