论文部分内容阅读
传统的定期维护制度成本高 ,劳动强度大 ,且对发动机故障的诊断和探测能力十分有限。现代飞机上的发动机监控系统 ( EMS)具有向维护人员提供有关发动机故障信息的潜在能力。本文将径向基函数 ( RBF)神经网络应用到航空发动机故障诊断中。该方法能够依靠测量参数探测发动机多个气路故障 ,并对各大部件的性能退化进行定量的诊断。仿真结果表明 ,诊断的精度能够满足实际应用的需要 ,神经网络的非线性映射能力可用来捕捉发动机的特性。该方法具有通用性 ,在其他类似的复杂机械中也可以获得应用。
The traditional periodic maintenance system is costly, labor-intensive, and has limited ability to diagnose and detect engine faults. Engine Monitoring Systems (EMS) on modern aircraft have the potential to provide maintenance personnel with information about engine failure. In this paper, radial basis function (RBF) neural network is applied to aero-engine fault diagnosis. This method can detect multiple gas path faults of the engine by measuring parameters and quantitatively diagnose the performance degradation of various components. The simulation results show that the diagnostic accuracy can meet the needs of practical application. The nonlinear mapping ability of neural network can be used to capture the characteristics of the engine. The method is versatile and can be used in other similar complex machines.