DocumentCode :
578355
Title :
The application of GA-based PID parameter optimization for the control of superheated steam temperature
Author :
Jing, Wang ; Yang, Xue ; Lei, Chen
Author_Institution :
Huaneng Shanghai Combined Cycle Power Co., Ltd., Shanghai, China
Volume :
3
fYear :
2012
fDate :
15-17 July 2012
Firstpage :
835
Lastpage :
839
Abstract :
Aiming at the characteristics as time-delay, large inertia of superheated steam temperature in the power plants, an optimal PID controller which takes the overshoot, rise time and setting time of the system as the performance indicators have been proposed. The PID parameters were optimized by the means of the genetic algorithm with real number encoding. Finally, a group of optimal PID parameters were obtained. Simulation results indicate the GA-Based PID controller in this paper still has the ability of self-learning and adaptive even the controlled object is greatly changed. The controller can acquire an ideal control effect with shorter dynamic transition time, smaller overshoot and oscillation when the controlled object is changing.
Keywords :
adaptive control; delays; genetic algorithms; learning systems; optimal control; performance index; steam; steam power stations; temperature control; three-term control; unsupervised learning; GA-based PID parameter optimization; adaptive control; dynamic transition time; genetic algorithm; optimal PID controller; oscillation; performance indicator; power plant; self-learning control; superheated steam temperature control; system overshoot; system rise time; system setting time; time-delay characteristics; Abstracts; Encoding; Genetics; Genetic algorithm; PID controller; Real coding; Superheated steam temperature;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2012 International Conference on
Conference_Location :
Xian
ISSN :
2160-133X
Print_ISBN :
978-1-4673-1484-8
Type :
conf
DOI :
10.1109/ICMLC.2012.6359461
Filename :
6359461
Link To Document :
بازگشت