一种基于神经网络的空空导弹传感器在线故障诊断技术研究

Research on Sensor Online Fault Diagnosis for Air-to-air Missile Based on Neural Network

为防止传感器故障影响空空导弹飞行任务达成,针对空空导弹非线性气动力模型参数难以确定的问题,提出了一种基于神经网络的传感器在线故障诊断方法。利用该方法构造了空空导弹加速度传感器的故障诊断器,并以数字仿真平台对常见的加速度传感器故障进行在线故障检测。仿真结果表明,设计的故障诊断器可有效诊断故障并准确估计传感器输出,具有一定工程应用价值。

In order to prevent sensor fault impact air-to-air missile flight mission accomplished,aiming at the parameter uncertainty for nonlinear aerodynamic model of air-to-air missile,a method for sensor online fault diagnosis based on BP neural network is proposed.A fault diagnosis for the acceleration sensor of air-to-air missile is constructed,which applies to the digital simulation,and analyzed the common fault of acceleration sensor online.The simulation results show that,the fault is effectively diagnosed and the sensor output is accurately estimated by the fault diagnosis,which has practical value.