电子镇流器故障诊断的变精度粗糙集模型*

南京大学学报(自然科学版) ›› 2010, Vol. 46 ›› Issue (5): 494–500.

电子镇流器故障诊断的变精度粗糙集模型*

赵荣泳 1 , 李翠玲 2 ** , 高晓康 3 , 王昭云 4

出版日期:2015-04-02 发布日期:2015-04-02
作者简介: (1. 同济大学 CIM S 研究中心, 上海, 200092; 2. 上海海事大学电气自动化系, 上海, 200135;
3. 上海应用技术学院机械与自动化工程学院, 上海, 200233; 4. 环球迈特照明电子有限公司, 上海, 201102)
基金资助:
上海市教委支出预算项目( 2008088) ,上海市重点学科建设项目( J50602) , 教育部留学回国人员科研启动基金

Fault diagnosis model based on variable precision rough set for electronic ballasts

Zhao Rong Yong 1 , Li Cui Ling 2 , Gao Xiao K ang 3 , Wang Zhao Yun 4

Online:2015-04-02 Published:2015-04-02
About author:(1. CIMS Research Center, T ongji University, Shanghai, 200092, China;
2. Department of Electrical Automation, Shanghai Maritime University, Shanghai, 200135, China;
3. School of Mechanical and Automation Engineering, Shanghai Institute of Technology, Shanghai, 200233, China;
4. Universal Lighting Technologies, Shanghai, 201102, China)

摘要/Abstract

摘要： 目前大批量电子镇流器在线质量检测环节中的故障诊断是典型的信息不确定性问题, 连续属性和离散属性并存, 故障根源间的关系复杂, 也无法建立故障根源与故障之间的精确数学模型. 为此, 本
文引入变精度粗糙集理论(VPRS) , 构建电子镇流器的智能故障诊断模型( VPRS?ID). 针对连续属性的离散化问题, 引入了自组织映射神经网络模型( SOM) , 自动的实现了连续属性的离散化; 针对训练集合
中论域对象的不一致问题, 提出了基于频度的论域对象更新模型; 针对部分诊断对象与已发现规则之间的不一致性问题, 通过合理控制变精度因子, 发现了隐藏在产品故障数据与故障根源之间的规则. 在
满足电子镇流器诊断精度要求的基础上, 发现了有用的诊断知识, 改善了经典粗糙集理论规则集合的适应性. 针对目前低效的电子镇流器的人工诊断方式, 进一步提出电子镇流器的在线智能诊断程序架构,
为准确和高效的自动化故障诊断, 提供决策支持. 最后, 通过电子镇流器故障诊断的工程实例, 验证了该智能诊断模型的有效性.

Abstract: In the mode of large-scale manufacture, the fault diagnosis of electronic ballasts for online quality testing is a typical uncertainty problem due to the complex relationship between the fault causes and the faults nowadays. A
precise mathematic model cannot be set up for the fault diagnosis of electronic ballasts. Thereby the variableprecision rough set theory is introduced to discover the rules between ballasts testing data and fault roots in this
paper. An intelligent diagnosis model based on variable precision rough set is established as VPRS-ID model. The self-organizing -mapping ( SOM) neural network is introduced in the clustering process for the continuous attributes
discretization. An update model based on object frequency is presented to solve the inconsistency between the training objects. T he variable precision factor is adjusted reasonably to solve the inconsistency between the some
testing objects and established rules in classic rough set theory. Some meaningful knowledge is discovered to improve the adaptability of classic rough set rules, based on the diagnosis precision requests of electronic ballasts. A
program structure of the online intelligent diagnosis for electronic ballasts is proposed to change traditional, low-efficient and manual diagnosis into automatic and intelligent diagnosis, acting as a decision support tool. Finally an
engineering sample illustrates the feasibility of this intelligent model.

赵荣泳 1 , 李翠玲 2 ** , 高晓康 3 , 王昭云 4
. 电子镇流器故障诊断的变精度粗糙集模型*
[J]. 南京大学学报(自然科学版), 2010, 46(5): 494–500.

Zhao Rong Yong 1 , Li Cui Ling 2 , Gao Xiao K ang 3 , Wang Zhao Yun 4. Fault diagnosis model based on variable precision rough set for electronic ballasts[J]. Journal of Nanjing University(Natural Sciences), 2010, 46(5): 494–500.

参考文献

[ 1 ] Kyumin C, Wonseok O, Jaeeul Y, et al. Electronic ballast of the electrode-less lamp for the group lighting system. International Conference on Electrical and Electronics Engineering, 2009, I?293?I ?296.
[ 2 ] Moo C S, Huang C K, Yang C Y. Acoustic resonance -free high?frequency electronic ballast for metal halide lamps. IEEE Transactions on Industrial Electronics, 2008, 55 ( 10 ) : 3653~ 3660.
[ 3 ] Gao X K, Li C L, Shi Y C, et al. A method of fault diagnosis of electronic ballast based on rough set theory. Low Voltage Apparatus, 2007, (24) : 34~ 38. ( 高晓康, 李翠玲, 施雨辰等. 基于粗糙集理论的电子镇流器故障诊断方法. 低压电器. 2007( 24): 34~ 38).
[ 4 ] Liu H X. Intelligent methods on information processing and their applications. Journal of Nanjing University ( Natural Sciences) , 2009, 45( 4) : 439~ 441. ( 刘红星. 信息处理的智能化方法及应用. 南京大学学报( 自然科学). 2009, 45( 4) : 439~ 441) .
[ 5 ] Cabrero -Canosa M, Hernandez-Pereira E, Moret?Bonillo V. Intelligent diagnosis of sleep apnea syndrome. Engineering in Medicine and Biology Magazine, IEEE, 2004, 23 ( 2 ) : 72~ 81.
[ 6 ] Han Y L, Zhang T H, Yang B R, et al. Intelligent diagnosis technology based on self-organizing. Computer Integrated Manufacturing Systems. 2007, 13(5): 1008~ 1014.
( 韩彦岭, 张桃红, 杨炳儒等. 基于自组织的智能诊断技术研究. 计算机集成制造系统, 2007, 13( 5) : 1008~ 1014) .
[ 7 ] Le Q H, Teng L, Zhu M Q, etal. On-line intelligent diagnosis and analysis system for quality control charts. Computer Integrated M anufac turing Systems, 2004, 10( 2): . 1583 ~ 1587.
( 乐清洪, 滕?霖, 朱名铨等. 质量控制图在线智能诊断分析系统. 计算机集成制造系统.2004, 10( 2) : 1583~ 1587).
[ 8 ] Pawlak Z. Rough set. International Journal of Information and Computer Science, 1982, 11 (5) : 341~ 356.
[ 9 ] Zhang W X. Rough set theory and method. Beijing: Science Press, 2001, 224. (张文修. 粗糙集理论与方法. 北京: 科学出版社, 2001, 224) .
[ 10] Gao X K. Research on intelligent diagnosis system based on rough set theory. Shanghai: Tongji University, 2007, 147. (高晓康. 粗糙集理论研究及其在工程和医学诊断中的应用. 上海: 同济大学, 2007, 147).
[ 11] Zhao R Y. Research on rough set theory, the digital factory technology and application in enterprise production DSS. Shanghai: Tongji University, 2005, 165. ( 赵荣泳. 粗糙集理论及数字化工厂技术在企业生产 DSS 中的应用研究. 上海: 同济大学, 2005, 165).
[ 12] Skowron A. Extracting laws from decision tables: A rough set approach. Computational Intelligence, 1995, (11): 371~ 388.
[ 13] Zhao R Y, Zhang H, Li C L, et al. The improvement of key algorithms for automatic rule generation in rough set theory. Computer Engineering and Applications, 2005, 41( 13): 32~ 35. ( 赵荣泳, 张 ?浩, 李翠玲等. Rough Set 规
则自动生成的关键算法改进. 计算机工程与应用, 2005, 41(13) : 32~ 35).

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed