基于IPSO-RVM的行星齿轮箱故障识别方法OA

To address the issue that the standard particle swarm optimization(PSO)algorithm is prone to falling into local optima when optimizing the kernel parameters of the relevance vector machine(RVM),which leads to insufficient accuracy of the planetary gearbox fault diagnosis model,a new fault identification method based on improved particle swarm optimization and relevance vector machine(IPSO-RVM)is proposed.The PSO algo-rithm is deeply improved by introducing nonlinear decreasing inertia weight and asymmetric adaptive learning factors to systematically balance its global exploration and local exploitation capabilities.The improved IPSO algorithm is used to adaptively search for the optimal kernel function parameters of the RVM model,thereby es-tablishing an IPSO-RVM intelligent diagnosis model.The proposed method is applied to fault diagnosis experi-ments using measured vibration signals from a planetary gearbox.The results show that the average classifica-tion accuracy of the proposed model reaches 92.55％,which has an improvement of 4.91％and 10.44％com-pared to the traditional PSO-RVM and PSO-SVM models,respectively.The experiments demonstrate that the proposed method effectively overcomes the problem of the PSO algorithm falling into local optima,finds better parameters for the RVM model,and significantly enhances the generalization ability and robustness of the diag-nosis model,providing a solution with higher accuracy and efficiency for the intelligent fault diagnosis of plane-tary gearboxes.