基于弹性小波神经网络的故障诊断研究

航空电源系统是机上设备的重要组成部分，任意一个环节出现故障，将会影响整个飞机系统的正常安全运行。针对神经网络收敛速度慢，易陷入局部最小的缺点，将小波神经网络结合弹性BP算法应用到电源系统故障诊断中。训练过程及仿真结果表明：小波神经网络故障诊断算法收敛时间方面表现更优，具有较高故障诊断率。

Research on Fault Diagnosis Based on Resilient Wavelet Neural Network

The aircraft electric power source system is an important part of the on board equipments, which will affect the normal and secure flight of the airplane. We combine the wavelet neural network with Rprop algorithm to diagnose the power system to settle the disadvantage of the neural network, which are the low convergence rate and easily running into local minimal. The training process and the result of simulation find out that the way we use has a faster convergence and a preferable diagnosis rate.