On-line genetic algorithm tuning of a PI controller for a heating system

Genetic algorithms have been used successfully for a number of off-line control optimisation problems. This paper investigates their use for the online regulation of temperature in a bench top heating system. Two approaches are considered: 1) using a time-invariant model of the plant dynamics with the GA adaptively choosing the PI controller parameters; 2) extending the idea using a time varying estimate of the plant model. Test results are given for both methods. The paper concludes that the approach offers good regulation and good promise for application to more complex multivariable systems.