Design of PID-type controllers using multiobjective genetic algorithms

被引:76
作者
Herreros, A [1 ]
Baeyens, E [1 ]
Perán, JR [1 ]
机构
[1] Univ Valladolid, ETSII, Inst Tecnol Avanzadas Prod, E-47011 Valladolid, Spain
关键词
PID controllers; multiobjective optimization; tuning methods; genetic algorithms; limits of performance;
D O I
10.1016/S0019-0578(07)60102-5
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The design of a PID controller is a multiobjective problem, A plant and a set of specifications to be satisfied are given. The designer has to adjust the parameters of the PID controller Such that the feedback interconnection of the plant and the controller satisfies the specifications. These specifications are usually competitive and any acceptable solution requires a tradeoff among them. An approach for adjusting the parameters of a PID controller based on multiobjective optimization and genetic algorithms is presented in this paper. The MRCD (multiobjective robust control design) genetic algorithm has been employed. The approach can be easily generalized to design multivariable coupled and decentralized PID loops and has been successfully validated for a large number of experimental cases. (C) 2002 ISA-The Instrumentation, Systems, and Automation Society.
引用
收藏
页码:457 / 472
页数:16
相关论文
共 24 条
[1]  
[Anonymous], GENETIC ALGORITHMS C
[2]  
[Anonymous], 1989, GENETIC ALGORITHM SE
[3]  
Astrom K. J., 1995, PID CONTROLLERS THEO
[4]   Design of PI controllers based on non-convex optimization [J].
Astrom, KJ ;
Panagopoulos, H ;
Hagglund, T .
AUTOMATICA, 1998, 34 (05) :585-601
[5]  
CANTUPAZ E, 1997, 95007 ILL GEN ALG LA
[6]   A GENETIC APPROACH TO MIXED H-2/H-INFINITY OPTIMAL PID CONTROL [J].
CHEN, BS ;
CHENG, YM ;
LEE, CH .
IEEE CONTROL SYSTEMS MAGAZINE, 1995, 15 (05) :51-60
[7]  
Chen BS, 1998, IEEE T CONTR SYST T, V6, P707, DOI 10.1109/87.726532
[8]  
Coello C. A. C., 1999, Knowledge and Information Systems, V1, P269
[9]  
DAKEV NV, 1995, P 1 IEE IEEE INT C G, P226
[10]  
Deb Kalyanmoy., 1999, EVOLUTIONARY ALGORIT, P135