Design and experimental evaluation of an evolutionary neural-net based PID controller

Author

Suzuki, Michiyo ; Katayama, Masaru ; Yamamoto, Toru

Author_Institution

Dept. of Technol. & Inf. Educ., Hiroshima Univ., Japan

Volume

1

fYear

2002

fDate

2002

Firstpage

260

Abstract

PID control schemes have been widely used for most industrial processes which are represented by nonlinear systems. This paper presents a new design scheme of evolutionary neural-net based PID controllers, whose connection weights are adjusted by using a real-coded genetic algorithm. The newly proposed control scheme is numerically evaluated on a simulation example, and experimentally evaluated on a pilot-scale temperature control system.

Keywords

control system synthesis; evolutionary computation; genetic algorithms; neurocontrollers; nonlinear control systems; optimal control; process control; three-term control; GA; connection weights; evolutionary neural-net based PID controller; industrial processes; nonlinear systems; pilot-scale temperature control system; real-coded genetic algorithm; Algorithm design and analysis; Control systems; Electrical equipment industry; Genetic algorithms; Industrial control; Nonlinear control systems; Nonlinear systems; Temperature control; Three-term control; Weight control;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Applications, 2002. Proceedings of the 2002 International Conference on

Print_ISBN

0-7803-7386-3

Type

conf

DOI

10.1109/CCA.2002.1040195

Filename

1040195