Intelligent Fault Diagnosis on Aircraft Full Electric Braking System Based on Improved ABC-RBF
Due to the problems of high average diagnostic error and long time consumption in existing fault diagnosis methods,an automatic fault diagnosis method for aircraft full electric braking system based on improved artificial bee colony radial basis function(ABC-RBF)neural network is designed.The design adopts a hardware architecture of"USB interface+ARM+FPGA",the fault signal collector is composed of an upper computer,signal attenuation circuit,etc.,it achieves fault signal acquisition for the aircraft's full electric braking system.Based on mutual information and variational mode decomposition(VMD),a signal denoising algorithm is used to denoise the collected signals.Using the improved ABC algorithm to optimize the parameters of the RBF neural network,ensu-ring the effectiveness of the optimization parameters.And the fuzzy sets are introduced to improve the performance of the network,the gradient descent method is adopted to update network training,and reduce the errors of diagnostic results.From this,with the denoised signal input,the trained RBF neural network is optimized to achieve automatic fault diagnosis of the aircraft full electric bra-king system.The results indicate that the deviation factor value of this method reaches a minimum of 0.08 × 10-3.The average diag-nostic iteration time for the three types of faults is relatively short,among which the average diagnostic iteration time for the main landing gear"walking"fault is the shortest.
fault signal collectorsignal denoisingimproved ABC-RBF neural networkaircraft full electric braking systemfault diagnosis
NETL
NSTL
万方数据
