Application of WEE and GA-SVM in the Classification of Reactor CRDM Current Fault
The reliability of the control rod drive mechanism(CRDM)determines the safety of the reactor,and effective monitoring of the control rod drive mechanism is extremely necessary.The coil current change can effectively reflect the operating status of the control rod drive mechanism.On the basis of wavelet energy value,the eigenvector of coil current based on wavelet energy entropy(WEE)is constructed by introducing sliding window and entropy theory,and the current fault classification algorithm of control rod drive mechanism based on support vector machine(SVM)is designed.The classification algorithm used genetic algorithm(GA)and particle swarm optimization(PSO)to optimize the penalty coefficient c and kernel function parameter g of the support vector machine,and accurately realized the classification of the current fault of the control rod drive mechanism.The results show that:(1)Compared with the wavelet energy value,the feature vector based on the wavelet energy entropy can better reflect the local char-acteristics of the coil current,and also more accurately realize the classification of the coil current fault;(2)Compared with the particle swarm optimization,genetic algorithm as a support vector machine optimization algorithm not only has accurate classifi-cation,but also has higher efficiency in parameter optimization.
CRDMCurrent MonitoringFault ClassificationWavelet Energy EntropySVMGA
万方数据
维普
