基于CPO-CNN-LSTM的起落架系统故障诊断方法研究  

Research on Fault Diagnosis Method of Landing Gear System Based on CPO-CNN-LSTM

作　　者：唐凌云 苏艳[1] 易子超 TANG Lingyun;SU Yan;YI Zichao(College of Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)

机构地区：[1]南京航空航天大学民航学院,江苏南京211106

出　　处：《测控技术》2025年第3期1-8,共8页Measurement & Control Technology

基　　金：国家重点实验室项目(SGNR0000KJJS2007673)。

摘　　要：起落架刹车系统是飞机的重要组成部分,及时准确地诊断起落架刹车系统的故障,可以避免因故障导致的事故,提高飞机安全性。针对起落架刹车系统现有诊断算法识别精度较低和缺乏系统的参数优化等问题,提出了一种利用冠豪猪优化器(Crested Porcupine Optimizer,CPO)算法优化卷积神经网络融合长短期记忆网络(Convolutional Neural Network-Long Short Term Memory,CNN-LSTM)的飞机起落架刹车系统故障诊断方法。利用CPO的快速寻优能力,将找到的最优参数代入CNN-LSTM中重新构建模型,对起落架飞参数据进行训练分类并输出结果。诊断实验中,以某型号飞机起落架刹车系统真实飞参数据为输入,对起落架刹车系统的常见故障模式进行分类。实验结果表明,所提出的故障诊断方法有较好的故障诊断性能和实际的应用价值。Landing gear braking system is an important part of aircraft.Timely and accurate diagnosis of fault in the landing gear braking system can avoid accidents caused by fault and improve the safety of aircraft.Aiming at the problems of low recognition accuracy and lack of system parameter optimization of existing diagnosis algorithms for landing gear braking system,a fault diagnosis method for aircraft landing gear braking system is proposed,which uses crested porcupine optimizer(CPO)algorithm to optimize convolutional neural networklong short term memory(CNN-LSTM).Using the fast optimization ability of CPO,the optimal parameters are substituted into CNN-LSTM to reconstruct the model,and the landing gear flight parameters are trained and classified and the results are output.In the diagnostic experiment,the real flight parameter data of the landing gear braking system of a certain type of aircraft is taken as input to classify the fault modes of the landing gear braking system.The experimental results show that the proposed fault diagnosis method has good fault diagnosis performance and practical application value.

关 键 词：起落架刹车系统 故障诊断 冠豪猪优化器算法 卷积神经网络 长短时记忆神经网络 

分 类 号：V19[航空宇航科学与技术—人机与环境工程]