Fault diagnosis of rectifier circuit based on DGWO-BVM algorithm

In order to quickly and accurately identify the thyristor faults in three-phase bridge rectifier circuits, this paper proposes a fault diagnosis model based on DGWO-BVM algorithm.Firstly, the fault voltage signal is decomposed in three layers by wavelet packet transform and normalized. The extracted feature vector is taken as the model input vector;The fault diagnosis model is constructed by using the advantages of the ball vector machine (BVM) in solving multi-classification problems.The parameters of the BVM model were selected by the Grey Wolf optimization algorithm (DGWO) which enhanced the leader's decision-making ability.Finally, the model is tested by the test sample set and compared with other algorithms. Experiments show that this model can improve the accuracy of fault diagnosis, reduce the time of diagnosis, and has strong matching and reliability.