论文部分内容阅读
提出了一种控制系统传感器故障诊断的新方法。该方法针对多传感器系统的特点，将相关传感器的输出数据综合，应用神经网络的非线性拟合能力扩展相关信息，综合信息诊断策略根据这些信息对故障传感器进行定位，同时对故障信号进行恢复。仿真实验表明该诊断方法对于传感器的常见故障能够进行识别和恢复，效果令人满意。
A new method for fault diagnosis of sensor in control system is proposed. According to the characteristics of multisensor system, this method synthesizes the output data of the relevant sensors and applies the nonlinear fitting ability of neural network to expand the related information. The comprehensive information diagnosis strategy locates the fault sensor according to the information and restores the fault signal. The simulation results show that this method can identify and recover the common faults of sensors, and the results are satisfactory.