基于SDAE的液压泵故障诊断方法研究  

Research on Fault Diagnosis Method for Hydraulic Pump based on SDAE

作　　者：吴智怀 王志伟 Wu Zhihuai;Wang Zhiwei(Beijing Metallurgical Equipment Research Design Institute Co.,Ltd.,Beijing 100029)

机构地区：[1]北京中冶设备研究设计总院有限公司,北京100029

出　　处：《冶金设备》2023年第5期15-22,共8页Metallurgical Equipment

摘　　要：为探究液压泵故障诊断的方法,将基于堆叠降噪自动编码器的图像识别算法引入到液压泵的故障诊断中。文章介绍了基于堆叠降噪自编码器的故障诊断方法,研究该方法在机械设备领域故障诊断的可行性,采用轴向柱塞液压泵的故障状态信号数据进行实验验证;首先采集液压泵不同状态下的振动信号进行预处理分析,进而构建故障诊断网络模型进行故障诊断,从实验结果可以看出,选择合适的时频预处理方法能够使得基于SDAE的液压泵故障诊断方法具有较高的准确率,采用主成分分析进一步表明了SDAE方法的特征提取能力,证明了该方法在液压泵的故障诊断的可行性。In order to explore the fault diagnosis method of hydraulic pump,the image recognition algorithm based on stack noise reduction autoencoder is introduced into the fault diagnosis of hydraulic pump.This paper introduces the fault diagnosis method based on stack noise reduction autoencoder,studies the feasibility of this method in the field of mechanical equipment fault diagnosis,and uses the fault state signal data of axial piston hydraulic pump to verify the experiment.Firstly,the vibration signals of hydraulic pumps under different states were collected for pre-processing and analysis,and then a fault diagnosis network model was constructed for fault diagnosis.It can be seen from the experimental results that selecting a suitable time-frequency pre-processing method can make the fault diagnosis method of hydraulic pumps based on SDAE have a high accuracy.The feature extraction capability of SDAE method was further demonstrated by principal component analysis,and the feasibility of this method in fault diagnosis of hydraulic pump was proved.

关 键 词：液压泵 故障诊断 SDAE 实验分析 

分 类 号：TH137[机械工程—机械制造及自动化] TP301[自动化与计算机技术—计算机系统结构]