Multi- Service Condition Load Monitoring Method for Landing Gear using BP Neural Networks

The landing gear loading experience in service conditions directly affects the aircraft takeoff and landing safety. In this work, a multi-condition load real-time monitoring method is proposed coupling fiber Bragg grating (FBG) sensors with Back Propagation (BP) neural network method. The finite simulation method (FEM) is used to analyze the landing gear loads, and the landing gear FBG deployment scheme is given. The BP neural network based landing gear multi-condition load monitoring method was proposed to solve the inaccurate accuracy for small loads problem of the traditional method. In addition, a loads ground calibration test platform for the landing gear is designed and constructed. The experiment results show that for the proposed method, the average relative prediction error of heading load is less than 4%, the average relative prediction error of lateral load is less than 4%, and the average relative prediction error of vertical load is less than 1%. Therefore, the load monitoring method proposed in this study, which integrates FBG sensors with the BP neural network, provides a solution for precise load monitoring under multiple working conditions.