基于变分模态分解与支持向量数据描述结合的液压泵性能退化评估方法  

Performance Degradation Assessment Method of Hydraulic Pump Based on Integrated VMD and SVDD

作　　者：韩可 姜万录[2,3] 雷亚飞[2,3] 张永顺[2,3] 张生 HAN Ke;JIANG Wanlu;LEI Yafei;ZHANG Yongshun;ZHANG Sheng(CRRC Nanjing Puzhen Co., Ltd., Nanjing Jiangsu 210031,China;Hebei Provincial Key Laboratory of Heavy Machinery Fluid Power Transmission and Control, Yanshan University, Qinhuangdao Hebei 066004,China;Key Laboratory of Advanced Forging & Stamping Technology and Science,Yanshan University, Ministry of Education of China, Qinhuangdao Hebei 066004, China)

机构地区：[1]中车集团南京浦镇车辆有限公司,江苏南京210031 [2]燕山大学河北省重型机械流体动力传输与控制重点实验室,河北秦皇岛066004 [3]燕山大学先进锻压成形技术与科学教育部重点实验室,河北秦皇岛066004

出　　处：《机床与液压》2019年第19期164-170,共7页Machine Tool & Hydraulics

基　　金：国家自然科学基金资助项目(51875498;51475405);河北省自然科学基金重点项目(E2018203339);河北省博士研究生创新资助项目(CXZZBS2018045)

摘　　要：针对液压泵性能退化过程定量评估,提出了变分模态分解(VMD)和支持向量数据描述(SVDD)相结合的综合评估方法。利用VMD方法将信号分解成一系列不同频率成分的BIMF分量,并用SVDD方法对异常点进行剔除;使用SVDD方法对正常状态样本进行训练得到超球体模型,并计算各样本到球心的距离;再将各样本到球心的距离转化为隶属度,作为性能退化指标。通过对轴向柱塞泵滑靴磨损和松靴故障实验数据分析,验证了该性能退化评估方法的有效性。In order to evaluate the performance degradation of hydraulic pump quantitatively,a comprehensive evaluation method of Variational Mode Decomposition(VMD)and Support Vector Data Description(SVDD)is proposed.Firstly,the signal was decomposed into Bidimensional Intrinsic Mode Function(BIMF)components of different frequency components by VMD method,and the outliers were removed by SVDD.Then,the SVDD method was trained by the normal samples,the hyper sphere model was obtained,and the distance from each sample to the center of the sphere was obtained.Then the distance from each sample to the center of the ball was transformed into membership degree,as the performance degradation index were uniformly described.Finally,analysis of experimental fault signals of slipper wear and loose slipper of hydraulic piston pump,the effectiveness of the performance degradation assessment method is fully verified.

关 键 词：变分模态分解 支持向量数据描述 液压泵 性能退化评估 

分 类 号：TH212[机械工程—机械制造及自动化] TH213.3