【摘要】 为适应民航系统发展的要求,确保飞机的飞行安全、提高飞机发动机的可靠性和利用率、改变对发动机故障的诊断和探测能力的局限性以及降低传统的定期维护成本,提出了应用自适应神经网络技术对飞机发动机进行诊断。该文根据在飞行中检测或地面维修时采集到的数据,并将这些数据分为故障现象和故障集分别进行编码,而后利用神经网络的自学习、联想、推测、记忆、容错、自适应和多模式处理等功能对飞机发动机的故障进行诊断。应用结果表明,诊断的精度基本符合实际应用的需要。更多还原

【Abstract】 For adapting requirement of developing civil aviation system,ensuring safety of airplane,improving reliability and utilization,changing the localization of aero engine fault diagnosis and exploring ability,and reducing traditional maintenance cost,this paper brings forward using an adaptive neural network to diagnose aero engine fault. The data gathered from flying check or ground maintenance is divided into two collections which are fault phenomena and fault points,and then is processed coding. Finally,it uses neural network functions that are self-study,associate,conjecture,memory,adaptive and so on to diagnose aero engine fault. The results obtained demonstrate that the accuracy of diagnosis is consistent with practical requirements.更多还原