Parameters estimation of continuous system using improved hybrid genetic algorithm

In order to improve the performance of genetic algorithm, a new hybrid genetic algorithm is provided firstly in this paper. It adds up the advantages of the genetic algorithm and gradient algorithm, and uses the results of gradient algorithm improving the populations of genetic algorithm, and selects the best value as the starting point of gradient algorithm next time by comparing the best individual of genetic algorithm with the last results of gradient algorithm. Parameter estimation of continuous system is finished by the improved hybrid genetic algorithm, and simulation results show it is more quickly than the simple genetic algorithm and owes better anti-noise ability; at the same time, it improves the defects of genetic algorithm with worse local searching ability, and it provides a new method for the parameters estimation of continuous system. 1548-7741/Copyright © 2008 Binary Information Press October 2008.