航空发动机燃油系统执行机构及其传感器故障诊断

提出了基于执行机构模型和航空发动机逆模型的执行机构及其传感器单一故障诊断和定位方法.基于执行机构小闭环结构建立了3阶执行机构传递函数模型.基于两个并联的BP(back propagation)神经网络,建立了航空发动机稳态逆模块和动态补偿模块,形成航空发动机逆模型,以实现基于航空发动机输出的燃油流量估计.以执行机构模型输出和传感器输出之间的偏差为依据进行故障判别,以航空发动机逆模型输出和传感器输出偏差为依据对故障进行定位.以某型航空发动机及其燃油系统执行机构模型为对象进行的仿真,结果表明,该诊断系统可在航空发动机稳态、动态情况下准确地诊断出幅值在1.6%以上的执行机构及其传感器故障并进行定位,验证了所提出故障诊断方法的有效性.

Fault diagnosis for actuator and its sensor of aero-engine fuel system

A single fault diagnosis and location method of aero-engine actuator and its sensor was proposed. The method was based on actuator model and aero-engine inverse model. A three-order transfer function model of actuator was established according to its closed-loop construction. The aero-engine inverse model was built based on two parallel BP (back propagation) neural network including a steady inverse module and a dynamic compensation module, so the fuel mass flow estimation can be achieved based on the outputs of aero-engine. Fault diagnosis according to the deviation between the actuator model output and the sensor output was adopted. Fault location according to the deviation between the aero-engine inverse model output and the sensor output was used. The simulation results on an aero-engine and its fuel system actuator models show that the fault diagnosis system can diagnose amplitude in more than 1.6% of actuator and its sensor fault and locate it accurately at the steady and dynamic states of aero-engine, which verifies the effectiveness of the proposed method.