[1]陆伟峰,朱庆保,崔红梅,等.基于ISODATA算法的自组织单输入单输出T-S模糊系统[J].南京师范大学学报(工程技术版),2006,06(01):061-66.
 LU Weifeng,ZHU Qingbao,CUI Hongmei.A Self-organizing Single-input Single-output T-S Fuzzy System Based on IOSDATA Algorithm[J].Journal of Nanjing Normal University(Engineering and Technology),2006,06(01):061-66.
点击复制

基于ISODATA算法的自组织单输入单输出T-S模糊系统
分享到:

南京师范大学学报(工程技术版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
06卷
期数:
2006年01期
页码:
061-66
栏目:
出版日期:
2006-03-30

文章信息/Info

Title:
A Self-organizing Single-input Single-output T-S Fuzzy System Based on IOSDATA Algorithm
作者:
陆伟峰;朱庆保;崔红梅;
南京师范大学数学与计算机科学学院, 江苏南京210097
Author(s):
LU WeifengZHU QingbaoCUI Hongmei
School of Mathematics and Computer Science,Nanjing Normal University,Nanjing 210097,China
关键词:
ISODATA算法 T-S模糊系统 PSO算法
Keywords:
ISODATA a lgo rithm T-S fuzzy system PSO algor ithm
分类号:
TP11
摘要:
T-S模糊模型已得到了广泛的研究与应用.但在该模型的建模过程中,在结构辨识、参数优化等方面仍存在一些不足,为此提出了一种基于ISODATA算法的自组织T-S模糊系统.该方法基于输入输出数据,分两步对模糊系统进行建模.第一步,使用基于线性原型的ISODATA算法,对输入输出数据进行聚类,确定系统结构.第二步,建立初始T-S模糊系统,然后使用粒子群算法优化系统参数.与传统方法相比,具有自动优化系统结构的优点.仿真结果验证了该方法的有效性.
Abstract:
T-S fuzzy model has been stud ied and applied w idely. Wh ile during the modeling prog ress of T-S fuzzy m ode ,l there‘ re som e problem s on structure identifica tion、param e ter optim ization etc. Th is pape r proposes a sel-f org an izing T-S fuzzy sy stem based on ISODATA algor ithm. From the input-output da ta, w em odel the fuzzy system during two steps. Step 1 gets optim a l c lass o f the input-output da ta using ISODATA a lgor ithm based on linear pro totype, and determ ines the system structure. Step 2 builds a sim ple T-S fuzzy model and optim izes the sy stem param eters by Pad ic le Swa rm Optim ization a lgor ithm. Sim ulation resu lts show the e ffectiv eness o f the propo sed m ethod.

参考文献/References:

[ 1] 王立新. 模糊系统: 挑战与机遇并存——十年研究之感悟[ J]. 自动化学报, 2001, 27 ( 4): 585-590.
WANG L ix in. Fuzzy system: opertun ity coex iting w ith challenge—— d igestion of ten-year research[ J]. A ctaAutom atica Sin-i ca, 2001, 27( 4): 585-590. ( in Chinese)
[ 2] WANG L X, M ENDEL JM. Fuzzy basis functions, un ive rsal approx im ation, and orthogonal least squares lea rning[ J]. IEEE T rans on Neural Netwo rks, 1992, 3( 5): 807-814.
[ 3] WANG L X, MENDEL JM. G enerating fuzzy rules by learn ing from ex am ples[ J]. IEEE T rans on Sy stem sM an and Cybern, 1992, 22( 6): 1414-1427.
[ 4] ABE S, LAN M S. Fuzzy rules ex traction directly from num erical data for function approx im ation[ J]. IEEE Trans Sy stM an Cybern, 1995, 25: 119-129.
[ 5] TAKAGI T, SUGENO M. Fuzzy identification o f system s and its applica tions to mode ling and contro l[ J]. IEEE Trans Syst M an Cybern, 1985, 15( 1): 116-132.
[ 6] ROJAS I, POMARES H, ORTEGA J, e t a.l Sel-f o rganized fuzzy sy stem generation from tra in ing ex am ples[ J]. IEEE T rans Fuzzy Syst, 2000, 8: 23-36.
[ 7] POMARES H, ROJAS I, ORTEGA J, et a.l A system a tic approach to a se l-f gene rating fuzzy rule-table for function approx im ation [ J]. IEEE T rans Sy stM an Cybe rn, 2000, 30: 431-447.
[ 8] HONG T, LEE C. Induction of fuzzy rules and membership functions from tra in ing ex am ples[ J]. Fuzzy Se ts Sy st, 1996, 84: 33-37.
[ 9] SH IGEO ABE, LAN M INGSHONG. Fuzzy ru les ex traction direc tly fo rm num er ica l data for function approx im a tion[ J]. IEEE T rans Sy stM an Cybe rn, 2000, 25( 1) : 119-129.
[ 10] WU TZUPING, CHEN SHY IM ING. A new m e thod for constructingm enbership functions and fuzzy rules from tra in ing ex amp les[ J]. IEEE Trans SystM an Cybern, Pa rt B: Cybe rnetics, 1999, 29( 1) : 25-40.
[ 11] ROJAS I, POMARES H, FEMANDEZ F J, et a.l A new m ethodology to obta in fuzzy sy stem s autonom ously from tra in ing data [ C ] / / IEEE Internationa l Fuzzy System s Confe rence Proceed ings FUZZ-IEEE’ 99. Seou:l [ s. n. ] , 1999, 1: 527-532.
[ 12] HECTOR POMARES, IGNACIO RO JAS, JESU’S GONZ IEZ, et a.l Structure identifica tion in com plete ru le-based fuzzy system s[ J]. IEEE T ransactions on Fuzzy System s, 2002, 10( 3): 349-359.
[ 13] JANETTE C. CANO, PATR IC IA A. Nava, a fuzzy me thod for automa tic generation of m embersh ip function using fuzzy relations from training exam ples [ C] / /Annua lM eeting of theNo rth Am er ican Fuzzy Inform a tion Processing So ciety ( NAFI PSFLINT 2002). New Orleans, 2002: 158-162.
[ 14] 李洪兴. 变论域自适应模糊控制器[ J]. 中国科学( E辑), 1999, 29( 1): 32-42.
LIH ongx ing. Se l-f adaptive fuzzy contro ller w ith chang ing un iverse o f d iscourse [ J]. Science in China ( Ser ies E) , 1999, 29( 1): 32-42. ( in Ch inese)
[ 15] SUDKAM P T, HAMMELL R J. Interpo lation, comp le tion and learn ing fuzzy rules[ J]. IEEE Trans SystM an Cybe rn, 1994, 24: 332-342.
[ 16] 边肇祺. 模式识别[M ]. 北京: 清华大学出版社, 2000: 234-241.
B IAN Zhaoq.i Pa ttern Recogn ition[M ]. Be ijing: Ts inghua University Press, 2000: 234-241. ( in Chinese)
[ 17] 高新波. 模糊聚类分析及其应用[M ]. 西安: 西安电子科技大学出版社, 2004.
GAO X inbo. Fuzzy C luster Ana ly sis and Its Applitions[M ]. X i’ an: Xidian University Press, 2004. ( in Ch inese)
[ 18] KENNEDY J, EBERHART R. Particle sw arm optim za tion[ C] / / Proceed ings of the IEEE Inte rnational Conference on Neural N etworks. 1995: 1942-1948.
[ 19] EBERHART R, KENNEDY J. A new optim ze r using partic le sw arm theory [ C] / / Proceedings o f the 6th In ternational Symposium onM icroM achine and H umm an Sc ience Pisca taw ay, Nagoy a: IEEE Serv ice Con ter, 1995: 39-43.
[ 20] 刘靖明, 韩丽川, 侯立文. 一种新的聚类算法?—— 粒子群聚类算法[ J]. 计算机工程与应用, 2005( 20): 183-185.
LIU Jingm ing, HAN Lichuang, HOU L iw en. A new cluster algorithm—— partic le swarm c luster a lgo rithn[ J]. Com puter Eng ineer ing and Applications, 2005( 20): 183-185. ( in Ch inese)

备注/Memo

备注/Memo:
作者简介: 陆伟峰( 1978-) , 硕士研究生, 主要从事智能算法与智能控制等方面的学习和研究. E-m ail:edw in- 7810@ 163. com
通讯联系人: 朱庆保( 1955-) , 教授, 主要从事机器人路径规划、人工智能与智能控制等方面的教学与研究.E-m ail: zhuqingbao@ n jnu. edu. cn
更新日期/Last Update: 2013-04-29