Fault diagnosis of electric rudder based on self-organizing differential hybrid biogeography algorithm optimized neural network

Abstract

Electric rudder is one of the most essential equipment on an aircraft. The diagnosis of electric rudder faults is of great significance to the production of electric rudder. To this end, this paper proposes a fault diagnosis method based on back-Propagation neural network (BPNN), which optimizes the weights and biases of BPNN using the self-organizing differential hybrid biogeography-based optimization algorithm (SODEBBO), and completes the automatic classification of multiple faults in electric rudder testing. The self-organizing differential hybrid biogeography-based optimization algorithm optimized the BPNN (SODEBBO-BP) model is compared with other eight models, and the results show that the maximum accuracy of SODEBBO-BP algorithm is 0.9918, the precision is 0.9981, F1-Score is 0.9981, G-mean is 0.9853, and Kappa is 0.9643 compared with other eight models. The model proposed in this paper is validated by experimental study and can be applied to the production and maintenance of electric rudder in the future.