基于观测器和神经网络的液压系统故障诊断方法研究  

Research on Fault Diagnosis Approach of Hydraulic System Based on Observer and Neural Network

作　　者：谢建[1] 何德虎[1] 李良[1] 

机构地区：[1]第二炮兵工程学院,陕西西安710025

出　　处：《机床与液压》2011年第1期135-137,共3页Machine Tool & Hydraulics

摘　　要：针对液压系统特点,提出基于观测器和神经网络的故障诊断方法。该方法的原理是基于观测器实现故障的判断,利用经观测器输出训练的神经网络实现故障定位。相对于故障树和专家系统等方法,该方法的优点是诊断速度更快、不需要大样本。仿真结果证明该方法有效。A fault diagnosis approach of hydraulic system based on observer and neural network was put forward aiming at the characteristics of hydraulic system. The principle of the approach was that the fault was judged by the observer and the position of the fault was confirmed by neural network which was trained by the output of the observer. The approach has a faster diagnosis speed than expert system and fault tree etc. It doesn＇t need a lot of samples. The simulation result demonstrates that the method is effective.

关 键 词：观测器 神经网络 液压系统 故障诊断 

分 类 号：TH137[机械工程—机械制造及自动化]