Improved Support Vector Machine for Circuit Breaker Fault Diagnosis Based on Spark Technology

In order to more quickly and accurately analyze the state and fault diagnosis of circuit breakers, this paper proposes a circuit breaker fault diagnosis method based on Spark technology with improved support vector machine. The method first uses independent component analysis to pre-process and screen out key data, which effectively reduces repetitive variables and reduces data complexity, and then improves the classification rate by combining with K-means clustering algorithm and establishes a K-means-SVM model, which then diagnoses the circuit breaker faults more accurately; and finally, it utilizes the support vector machine constructed in the Spark machine learning Finally, the support vector machine model constructed in Spark machine learning is used for parallel data analysis and processing, and the relevant performance indexes are obtained through the rapid evaluation of the model, and the classification and diagnosis results are outputted with the optimal model of K-means-SVM, which greatly improves the diagnosis efficiency through the parallelization of the computation in Spark. The results show that the method is fast in computation and can effectively solve the problem of few training samples, and at the same time has a better diagnostic effect.