基于BP神经网络的工程机械液压系统自动故障诊断研究  

Research on Automatic Fault Diagnosis of Construction Machinery Hydraulic System Based on BP Neural Networks

作　　者：曾行健 丁悦 汤清源 白龙 ZENG Xingjian;DING Yue;TANG Qingyuan;BAI Long(School of Mechanical Science&Engineering,Huazhong University of Science and Technology,Wuhan Hubei 430074,China;Jianghan Machinery Research Institute Limited Company of CNPC,Jingzhou Hubei 434000,China)

机构地区：[1]华中科技大学机械科学与工程学院,湖北武汉430074 [2]中石油江汉机械研究所有限公司,湖北荆州434000

出　　处：《机床与液压》2025年第2期165-170,共6页Machine Tool & Hydraulics

基　　金：国家重点研发计划项目(2022YFB4700300);湖北省重点研发计划项目(2023BAB067)。

摘　　要：工程机械液压系统由于结构复杂而不可避免地出现故障,常规检测方法为人工检测,但检测过程费时费力。针对此问题,对工程机械液压系统进行建模及简化,并结合BP神经网络学习故障数据。在管路系统中安装压力和流量监测仪以跟踪数据,通过调整选定元件的参数来模拟故障情况,记录监测仪的数据并进行整理,采用主成分分析法进行信息抽取和降维,并作为神经网络的输入。同时,手动标注元器件的当前状态作为训练标签。对每个元件均构建了一个独立的神经网络模型,用于学习输入数据与标签之间的关系。结果显示:阀和泵的准确率分别达到98.61%和96.52%,表明模型的准确率较高,可实现工程机械液压模型的自动故障诊断。The faults of the construction machinery hydraulic system are inevitable due to its complex structure,the routine detection method is manual detection,but the detection process is time-consuming and labor-intensive.In view of this problem,the construction machinery hydraulic system was modeled and simplified,and combined with the BP neural network to learn the fault data.Pressure and flow monitors were installed in the pipeline system to track operational data.Fault conditions were simulated by modifying the parameters of selected components,and the data from these monitors were recorded and processed.Principal component analysis(PCA)was employed to extract information and reduce dimensionality for using as input to neural networks.Additionally,the current operational status of the components was manually labeled to serve as the training labels.For each component,a separate neural network was constructed to learn the relationship between its input data and the corresponding labels.The results show that the accuracies of the valve and the pump reach 98.61%and 96.52%,respectively,indicating the high accuracy of the model.So automatic fault diagnosis of construction machinery hydraulic model can be realized.

关 键 词：BP神经网络 故障诊断 工程机械液压系统 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]