Fault Diagnosis of Turnouts Based on VMD and GA-BPNN

To address the issue of low efficiency in switch fault diagnosis, a fault diagnosis method is proposed that combines Variational Mode Decomposition (VMD) and Genetic Algorithm (GA) to optimize the Back Propagation Neural Network (BPNN). First, the power curves under various operating states of the switch are analyzed, and VMD is used to decompose the fault signal into a series of intrinsic mode functions (IMFs). Then, the permutation entropy of each state signal's IMF is calculated as the fault feature vector. Finally, the fault data is divided into training and testing sets in a 7:3 ratio, and the data is input into the GA-LSSVM model for training and evaluation to obtain the optimal fault diagnosis model. The experimental results show that the fault diagnosis method proposed in this paper achieves an accuracy of 86.25%, a recall rate of 88.42%, and an F1 score of 0.92, significantly improving the fault diagnosis efficiency of the switch.