›› 2013, Vol. 26 ›› Issue (4): 33-.

• 论文 • 上一篇    下一篇

基于CBR循环发电机组故障诊断系统研究

聂建华,丁琳,酆峰   

  1. (1.安徽工业大学 电气信息学院,安徽 马鞍山 243000;2.济钢集团公司 能源动力厂,山东 济南 250132)
  • 出版日期:2013-04-15 发布日期:2013-06-20
  • 作者简介:聂建华(1962—),男,硕士,教授。研究方向:智能仪表、流程工业过程控制,数控系统应用。丁琳(1989—),女,硕士研究生。研究方向:故障诊断。E-mail:dinglin13637101117@126.com。酆峰(1979—),男,硕士。研究方向:电子技术应用。

Research into Cycle Generating Unit Fault-diagnosis System Based on CBR Technology

 NIE Jian-Hua, DING Lin, FENG Feng   

  1. (1.School of Electrical Information,Anhui University of Technology,Ma'anshan 243000,China;
    2.Energy Power Plant,Jinan Iron and Steel Group Company,Jinan 250132,China)
  • Online:2013-04-15 Published:2013-06-20

摘要:

针对某钢铁公司循环发电机组的故障诊断问题,应用CBR技术进行故障诊断系统的设计。提出基于VSM的相似度分类器策略完成CBR关键技术的设计。通过计算特征项权重用向量表示机组故障案例,采用余弦法计算案例间的相似度,比较输出相似度大于阈值的案例。最终在asp.net平台上开发了基于Web的故障诊断系统。案例分析结果表明,基于CBR故障诊断系统能提高新故障维修解决的准确性和高效性。

关键词: CBR, 故障诊断, 循环发电机组, 相似度

Abstract:

In view of the fault diagnosis problem of cycle generating unit in a steel company,a fault diagnosis system based on CBR technology is designed.VSM (vector space model) similarity classifier is adopted to complete the design of the key technology of CBR.Fault cases are represented by vectors through calculating the weight of feature items;the similarity between the cases is calculated by cosine method;and the cases whose similarity is greater than the threshold value are output.Finally,the fault diagnosis system is developed based on WEB on ASP.NET platform.Case analysis results show that the proposed fault diagnosis system based on CBR can improve the accuracy and efficiency.

Key words: CBR;fault diagnosis;cycle generating unit;similarity

中图分类号: 

  • TP181