A hybrid of electromagnetism-like mechanism and back-propagation algorithms for recurrent neural fuzzy systems design

被引:28
作者
Lee, Ching-Hung [1 ]
Chang, Fu-Kai [1 ]
Kuo, Che-Ting [1 ]
Chang, Hao-Hang [1 ]
机构
[1] Yuan Ze Univ, Dept Elect Engn, Tao Yuan, Taiwan
关键词
electromagnetism-like algorithm; neural fuzzy system; nonlinear control; identification; mobile robot; OPTIMIZATION; NETWORK; MODEL; CONVERGENCE; CONTROLLER; PREDICTION;
D O I
10.1080/00207721.2010.488758
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This article introduces a novel hybrid evolutionary algorithm for recurrent fuzzy neural systems design in applications of nonlinear systems. The hybrid learning algorithm, IEMBP-improved electromagnetism-like (EM) with back-propagation (BP) technique, combines the advantages of EM and BP algorithms which provides high-speed convergence, higher accuracy and less computational complexity (computation time in seconds). In addition, the IEMBP needs only a small population to outperform the standard EM that uses a larger population. For a recurrent neural fuzzy system, IEMBP simulates the 'attraction' and 'repulsion' of charged particles by considering each neural system parameters as a charged particle. The EM algorithm is modified in such a way that the competition selection is adopted and the random neighbourhood local search is replaced by BP without evaluations. Thus, the IEMBP algorithm combines the advantages of multi-point search, global optimisation and faster convergence. Finally, several illustration examples for nonlinear systems are shown to demonstrate the performance and effectiveness of IEMBP.
引用
收藏
页码:231 / 247
页数:17
相关论文
共 44 条
[11]   Immune algorithm [J].
Ge, H ;
Mao, ZY .
PROCEEDINGS OF THE 4TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-4, 2002, :1784-1788
[12]  
Goldberg DE., 1989, GENETIC ALGORITHMS S, V13
[13]  
GORRINI V, 1994, PROCEEDINGS OF THE THIRD IEEE CONFERENCE ON FUZZY SYSTEMS - IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE, VOLS I-III, P193, DOI 10.1109/FUZZY.1994.343687
[14]   Comparison of particle swarm optimization and backpropagation as training algorithms for neural networks [J].
Gudise, VG ;
Venayagamoorthy, GK .
PROCEEDINGS OF THE 2003 IEEE SWARM INTELLIGENCE SYMPOSIUM (SIS 03), 2003, :110-117
[15]   ON FUZZY MODELING USING FUZZY NEURAL NETWORKS WITH THE BACKPROPAGATION ALGORITHM [J].
HORIKAWA, S ;
FURUHASHI, T ;
UCHIKAWA, Y .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (05) :801-806
[16]   A Hybri of genetic algorithm and particle swarm optimization for recurrent network design [J].
Juang, CF .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2004, 34 (02) :997-1006
[17]   A TSK-type recurrent fuzzy network for dynamic systems processing by neural network and genetic algorithms [J].
Juang, CF .
IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2002, 10 (02) :155-170
[18]  
Kim DH, 2002, PROCEEDINGS OF THE 2002 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS, VOL 1 & 2, P408, DOI 10.1109/FUZZ.2002.1005025
[19]  
Lee C.H., 2009, P INT MULT ENG COMP, V1, P224
[20]  
Lee C.H., 2007, 2007 CACS INT AUT CO