基于一种新型神经网络的导弹姿控故障诊断

Application of the New Neural Network in the Fault Diagnosis of Controlled System of Missile Posture

文中从改进网络结构出发,首先介绍了带偏差单元的递归神经网络的结构及算法,并将它和一改进算法的BP(以后称为FBP)网络分别用来对同一导弹姿态控制系统进行故障诊断。结果表明,该网络能很好地克服BP网络收敛速度慢等缺点,故障诊断的正确率优于该FBP神经网络。由于结构特点,它的应用范围比BP网络更加广泛。

This paper introduces the deviation units recurrence neural network and its algorithm.Then,the deviation units recurrence neural network and network based on an improved BP algorithm(FBP)respectively are presented to diagnose faults in the same Controlled System of Missile Posture.The test results demonstrate it can overcome the slow convergence of BP network etc.and is more precise than FBP network.The results also show that the range of its application is more extensive than BP network due to its structur...