多传感器信息融合技术在液压系统故障诊断中的应用  

Application of Multi-sensor Information Fusion for Hydraulic System Fault Diagnosis

作　　者：任凤娟 REN Feng-juan(Department of Mechanical and Electrical Engineering,City University of Zhengzhou,Zhengzhou 452370,China)

机构地区：[1]郑州城市职业学院机电工程系

出　　处：《液压气动与密封》2019年第7期52-55,共4页Hydraulics Pneumatics & Seals

基　　金：河南省重大科技专项(161100210600)

摘　　要：针对单参数诊断复杂系统中出现的信息不完整和不确定性问题,提出基于BP神经网络和D-S证据理论的多传感器信息融合故障诊断方法。为了简化BP神经网络结构,首先利用两并行BP神经网络对故障数据进行诊断;之后,D-S通过证据理论融合局部诊断结果,实现对于不准确信息的准确判断,获得准确诊断结果。该方法适用于特定类型火箭发射器液压驱动伺服系统(HDSS)的故障诊断,实现了对于液压伺服驱动系统中主要部件的故障定位和诊断,有效提高了系统可靠性。In order to solve the problem of incomplete and uncertain information in a single parameter diagnosis complex system,a multi sensor information fusion fault diagnosis method based on BP neural network and D-S evidence theory is proposed.In order to simplify the BP neural network structure,the two parallel BP neural network is used to diagnose the fault data.After that,the local diagnosis results are fused by the evidence theory to realize accurate inference of inaccurate information and obtain accurate diagnosis results.This method is suitable for the fault diagnosis of a specific type rocket launcher hydraulic drive servo system(HDSS).The fault location and diagnosis of the main components in the hydraulic servo drive system are realized,and the reliability of the system is effectively improved.

关 键 词：信息融合 D-S证据理论 BP神经网络 故障诊断 液压系统 

分 类 号：TH137[机械工程—机械制造及自动化]