Comparison of three optimisation methods for constrained generalised predictive control coupled with an estimation by genetic algorithm

This paper presents an application of the dichotomy and Gradient Method to constrained generalised predictive control. Using a constraints reduction method developed elsewhere, we show how the dichotomy algorithm and the Gradient Method can be applied to solve the constrained generalised predictive control efficiently and with less computation burden than that of other methods. It is also shown how the dichotomy method could be applied to system identification but with a higher computation burden than that of the recursive least squares method. It is also shown that Gradient Method combined with Genetic algorithm for system identification can give good results.