基于灰色神经网络的航空发动机振动故障诊断方法研究

针对某型双转子航空涡扇发动机台架试车中出现的振动异常现象,提出了一种基于灰色神经网络的航空发动机故障诊断方法,该方法不仅具有灰色关联度故障诊断方法计算量小、准确率高等优点,同时具备了神经网络强大的并行处理能力和良好的容错性。实例证明,该方法能够简单有效地诊断出航空发动机台架试车中出现的常见故障,为航空发动机故障诊断提供了一个新的途径。

The research on areoengine vibration fault diagnosis using grey neural network

A method of aviation engine fault diagnosis Based on grey neural network is proposed in this paper against the vibration anomalies in a dual-rotor aviation turbofan engine test bench.This method not only has the advantage of small calculation and high accuracy using Grey correlation method of fault diagnosis,but also has powerful parallel processing capability and good fault-tolerant of neural network.The examples demonstrate that this method can diagnose the common fault in the aeroengine test bench simply and effectively.This new method provides a new approach for the fault diagnosis of aeroengine.