FSVM模糊隶属度改进和信息熵融合的复杂设备故障分析  

Method for Fault Diagnosis and Identification of Complex Equipment Based on Improved FSVM and Fuzzy Membership Method Combined with the Information Entropy

作　　者：王宇凡[1] 董仲慧[2] WANG Yufan;DONG Zhonghui(School of Economics and Management,Xi’an Technological University,Xi’an 710021,China;School of Management,Northwestern Polytechnical University,Xi’an 710072,China)

机构地区：[1]西安工业大学经济管理学院,西安710021 [2]西北工业大学管理学院,西安710072

出　　处：《西安工业大学学报》2017年第11期794-801,共8页Journal of Xi’an Technological University

基　　金：国家自然科学基金项目(71502133);陕西省社会科学基金项目(2015R018);西安工业大学校长基金(XAGDXJJ15017)

摘　　要：为了实现更好的复杂设备故障诊断和识别,文中提出了基于FSVM模糊隶属度改进和信息熵融合的复杂设备故障诊断和识别的分析技术.对标准FSVM的模糊隶属度函数改良后建立了基于数据特征分布的高斯隶属度计算模型.通过工况数据验证了改良模糊隶属度与信息熵融合的FSVM对于复杂设备的故障诊断识别和工况状态评估更加有效.计算出设备故障模式与故障原因之间的权值,建立了一个多参数的复杂设备状态分析模型;并对各类故障原因对设备整体状态的影响进行定量分析,为复杂设备故障诊断和识别提供量化参考指标.In order to effectively diagnose and identify the faults of complex equipment,a method for fault diagnosis and identification is proposed based on the improved fuzzy support vector machine(FSVM)and fuzzy membership method combined with the information entropy technique.A calculation model of new gaussian fuzzy membership was established based on the data characteristics distribution after the standard fuzzy support vector machine(FSVM)and fuzzy membership method had been improved.Tests were conducted to verify that the proposed method was more effective in the fault diagnosis and identification.The weighted values between fault modes and fault causes were determined and an equipment performance analysis model was established,by which the effects of fault causes on the overall performance of complex equipment were analyzed quantitatively.The research provides a quantitative reference index for fault diagnosis and identification.

关 键 词：模糊支持向量机 模糊隶属度 信息熵 故障诊断识别 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]