提出了一种基于多重结构神经网络的故障检测方法。针对以歼击机为代表的非线性系统中存在的突发结构故障 ,构造了一个多重结构神经网络 ,在输入层对残差信号进行二进离散小波变换 ,提取其在多尺度下的细节系数作为故障特征向量 ,并将其输入到神经网络分类器进行相应的模式分类 ,之后再利用下一级的故障度辨识神经网络对故障的大小进行辨识。仿真结果表明 ,本文方法为歼击机组合结构故障的检测提供了有效的方法和途径。 A fault detection method based on multiple structure neural networks is proposed. Aiming at the sudden structural failure in the nonlinear system represented by fighter plane, a multi-structure neural network is constructed. The residual signal is subjected to binary discrete wavelet transform at the input layer, and its detail coefficient at multi-scale is extracted as fault Feature vectors and input them into the neural network classifier to carry on the corresponding pattern classification, and then use the next level of fault degree identification neural network to identify the size of the fault. The simulation results show that the method proposed in this paper provides an effective method and method for the detection of composite structural faults in fighter aircraft.