摘要
为提高液压机故障诊断的准确率,针对液压机故障诊断与性能评估困难、模型难以建立、智能化故障诊断装置缺乏的现状,文章提出了HSMM与SVM的混合算法。通过分析不同模式下的分类精度、响应速度与稳定性,来揭示特征数据提取与算法设计的内在联系,以此构建故障诊断的HSMM-SVM模型,并验证该模型对故障诊断的准确度,以及对信息缺失数据的鲁棒性。并以某液压机的故障诊断为例,验证了在HSMM、SVM、HSMM-SVM下的故障分类精度,说明HSMM-SVM对故障分类的有效性与适用性更好。
In order to improve the accuracy of fault diagnosis for large hydraulic press,a hybrid algorithm of HSMM and SVM is proposed to analyze the classification accuracy,response speed and stability under different modes and to reveal the characteristic number in view of the difficulties in fault diagnosis and performance evaluation,model establishment and lack of intelligent fault diagnosis devices for large hydraulic press.The intrinsic relationship between data extraction and algorithm design.The HSSMM-SVM model for fault diagnosis is constructed,and the fault diagnosis accuracy of the model is verified,which is robust to missing data.Finally,the fault diagnosis of a hydraulic press is taken as an example to verify the accuracy of fault classification under HSMM,SVM and HSMM-SVM.It shows that the classification under HSMM-SVM has more validity and applicability.
作者
何彦虎
HE Yan-hu(School of Mechanical and Electrical Engineering,Huzhou Vocational and Technological College,Huzhou 313000,China)
出处
《湖州职业技术学院学报》
2019年第4期57-62,共6页
Journal of Huzhou Vocational and Technological College
基金
2019年度浙江省科技厅公益项目“大型液压机故障诊断与性能评估关键技术及装置研发”(LGG19E050005)的研究成果之一。
关键词
液压机
故障诊断
HSMM-SVM
马尔科夫
hydraulic press
fault diagnosis
Hidden Semi-Markov Model-Support Vector Machine
Markov