BP神经网络在自行火炮电启动系统故障诊断中的应用  

Application of BP neural network in fault diagnosis of electric starting system in self-propelled gun

作　　者：谢正喜[1] 廖学兵[1] 赵海燕[1] 吴建磊[1] 

机构地区：[1]武汉军械士官学校自行火炮系,湖北武汉430075

出　　处：《移动电源与车辆》2007年第4期29-31,35,共4页Movable Power Station & Vehicle

摘　　要：自行火炮电启动系统的故障征兆与故障原因之间具有很强的非线性映射关系,通常对其进行在线故障诊断是比较困难的。利用BP神经网络的非线性映射的特性,由神经网络来学习及存储自行火炮电启动系统的故障征兆信号和故障原因之间的映射关系,并将其用于在线诊断,从而达到对自行火炮电启动系统的故障进行在线自动故障诊断的目的,极大地提高了故障的诊断效率。In the electric starting system of the self-propelled gun,because of the existence of non-liner mapping relationship between fault appearance and fault reason, on-line fault diagnosis is very difficult. This paper makes use of the non-linear mapping character of BP nerural network. Because it can study and save the relationship between fault appearance and fault reason, so we use it in the on-line diagnosis of electric starting system in self- propelled gun. And in this way, the efficiency of fault diagnosis can be raised extremely.

关 键 词：BP神经网络 自行火炮 电启动系统 故障诊断 

分 类 号：TN711.1[电子电信—电路与系统]