An efficient fusion approach to rule extraction based on rough set theory and particle swarm optimization and its application

作     者：Deng, Wu Yang, Xinhua Zou, Li Zhao, Huimin Li, Wen 

作者机构：Dalian Jiaotong Univ Software Inst Dalian 116028 Peoples R China Chongqing Univ State Key Lab Power Transmiss Equipment & Syst Se Chongqing Peoples R China Anhui Univ Key Lab Intelligent Comp & Signal Proc Hefei Peoples R China Dalian Univ Key Lab Adv Design & Intelligent Comp Dalian Peoples R China Sichuan Univ Sci & Engn Artificial Intelligence Key Lab Sichuan Prov Chengtu Peoples R China 

出 版 物：《PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART I-JOURNAL OF SYSTEMS AND CONTROL ENGINEERING》 (机械工程师学会会报；I辑：系统与控制工程杂志)

年 卷 期：2012年第226卷第I7期

页      面：904-913页

核心收录：

学科分类：08[工学] 0802[工学-机械工程] 0811[工学-控制科学与工程] 

基　　金：State Key Laboratory of Power Transmission Equipment & System Security and New Technology (Chongqing University) [2007DA105 12711406] National Natural Science Foundation of China National High Technology Research and Development Program of China (863 Program) [2012AA040912] Key Laboratory of Intelligent Computing & Signal Processing, Ministry of Education (Anhui University), China Artificial Intelligence Key Laboratory of Sichuan Province (Sichuan University of Science and Engineering), China [2010RZ004, 2011RYJ03] Key Laboratory of Advanced Design and Intelligent Computing (Dalian University), Ministry of Education, China [ADIC2010008] Key Laboratory of Numerical Simulation in the Sichuan Provincial College (Neijiang Normal University), China [2011SZFZ001] Traction Power State Key Laboratory of Southwest Jiaotong University [TPL1203] 

主　　题：Rule extraction rough set theory particle swarm optimization incomplete fault information centralized control substation fusion approach 

摘      要：Rule extraction is viewed as an important pretreatment step for machine learning and data mining. In allusion to the shortcomings of the rough set method for rule extraction in systems with incomplete fault diagnosis, in this paper we propose a novel fusion approach based on rough set theory and particle swarm optimization for rule extraction (RSTPSORE) in order to improve the diagnostic robustness and accuracy. In the proposed method, first, an iterative linear subsection interpolation completion method is used to achieve completion of the fault diagnosis process. Then rough set theory is used to find a subset which can preserve the meaning of the attributes. Particle swarm optimization is used to discover the best rule within the flying subset space. To verify the efficiency of the proposed method, experiments are carried out on the collected data in a centralized control substation incomplete fault diagnosis system. The results indicate that the proposed method can provide an efficient solution to finding a minimal subset.