基于LS-SVM液压缸泄漏故障诊断方法的研究  被引量:5

Research on Leakage Fault Diagnosis Method of Hydraulic Clinyder Based on LS-SVM

在线阅读下载全文

作  者:廖辉[1,2] 乔东凯[1] LIAO Hui QIAO Dongkai(School of Mechatronics, Guangdong University of Petrochemical Technology, Maoming Guangdong 525000, China Development Center of Technology for Fruit & Vegetables Storage and Processing Engineering, Guangdong Higher Education, Maoming Guangdong 525000, China)

机构地区:[1]广东石油化工学院机电学院,广东茂名525000 [2]广东高校果蔬加工与贮藏工程技术开发中心,广东茂名525000

出  处:《机床与液压》2017年第15期184-187,共4页Machine Tool & Hydraulics

基  金:广东省科技计划资助项目(2014A010106033);广东省自然科学基金博士启动项目(2014A030310376);广东省普通高校青年创新人才项目(631051)

摘  要:液压系统的执行机构是液压缸,其运动故障主要是压力不足,导致其动作反应不及时,针对液压控制系统中出现故障复杂性的特点,以液压系统的液压缸泄漏为研究对象,通过分析液压缸泄漏程度的故障形式,通过液压缸泄漏程度的分类,提出了基于LS-SVM液压缸泄漏故障方法,建立了该系统的故障诊断模型,对液压缸泄漏量进行定量的分析和算法研究。研究表明,这种针对液压缸泄漏故障诊断的原理、方法和特点,可以比较准确地预测液压缸泄漏的程度,对其故障诊断是有效和实用的。The cylinder is the executing agency of hydraulic system, the main performance of the motion fault of which is power shortage, which will cause its action is not timely and soon. According to the complex characteristics of fault appeared in the hydraulic control system, the hydraulic cylinder leakage is researched as an object. The analysis was carried out of the fault form in leakage de- gree of hydraulic cylinder, and then the leakage degree was classfied, the leakage fault diagnosis method of hydraulic clinyder was pro- posed based on LS-SVM, and fault diagnosis model of the system was established. Quantative analysis and alogrithm research were car- ried out on the leakage rate of cylinder. Researches show that this method of the hydraulic cylinder leakage fault diagnosis aimed at principle, method and characteristics can be more accurate to predict the degree of hydraulic cylinder leakage, its fault diagnosis is ef- fective and practical.

关 键 词:液压系统 SVM故障诊断 油液泄漏 

分 类 号:TK515[动力工程及工程热物理—热能工程] TP273.4[自动化与计算机技术—检测技术与自动化装置]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象