Application of multidimensional feature-based and optimized SVM in high-voltage circuit breaker fault classification
Aiming at the use of current signals for high-voltage circuit breaker fault classification process,the acquisition of current signals raw feature extraction of a single type,low fault recognition rate and degradation of classification performance,this paper proposes a fault classification method based on the combination of multidimensional features and support vector machine(SVM).Firstly,the critical time of the tripping current signal and the current amplitude are extracted as local features,and the global features of the current signal are extracted to form a multi-dimensional feature vector,which constructs a joint original feature set of the current of the circuit breaker operation process.Secondly,in order to eliminate redundant feature information,the final set of feature vectors is constructed after dimensionality reduction using principal component analysis(PCA).Finally,particle swarm algorithm(PSO)is used to optimize the support vector machine parameter setting problem for fault classification of circuit breakers.The experimental results show that the recognition accuracy is high using the method proposed in this paper,which has practical engineering application value.
万方数据
维普
