|
OPTIMAL PID TUNING WITH GENETIC ALGORITHMS FOR NON LINEAR PROCESS MODELS
J.M. Herrero, X. Blasco, M. Martínez, J.V. Salcedo
Predictive Control and Heuristic Optimization Group
Department of Systems Engineering and Control
Universidad Politécnica de Valencia
Camino de Vera 14, P.O. Box 22012 E-46071 Valencia, Spain
Tel: +34-963877000 ext: 5713. Fax: +34-963879579.
E-mail: xblasco@isa.upv.es http://ctl-predictivo.upv.es
This work presents a powerful and flexible alternative for tuning PID controllers using Genetic Algorithms. The potential of this technique is shown using non-linear process models and a reference trajectory. Flexibility is demonstrated by showing how to tune an optimal PID in various situations: model errors, noisy input, IAE minimization, and following a reference models, etc. These problems are solved by changing the minimization index.
Keywords: Genetic Algorithms, Optimization, PID, Robust Control, Non-linear Control
Session slot T-Th-M16: Evolutionary Algorithms in Control Design/Area code 2a : Control Design
 | 