在随动系统传统检测方法的基础上,提出一种采用神经网络技术的故障诊断方法。利用随动适配器在检测过程中故障的特征参数,采用神经网络技术分别对每一类故障进行诊断,网络的输入是与输出故障相关联的监测信号的特征值,对各网络输出进行决策判断,最后给出诊断结果。仿真实例结果表明该方法通过利用检测的各种特征信息,能够有效的诊断故障,提高诊断的准确性。

It puts forward a method of using neural network to detect fault for the servo adaptor based on tradition detection method. Utilizing the fault characteristic parameters in the course of detecting, it is developed for the method of using the neural network to diagnose every kind of fault. The inputs of network are the characteristic values of detecting signals related with the fault, and the outputs of network are judged to obtain final the result. Simulation examples show that the method of neural network can make the best of characteristic information of detecting signals to obtain effective diagnosis result, and improve the precision of diagnose.