The Investigation Of Control System In Parameter Change And Exterior Disturb Based On Genetic Algorithm And H Control

The system circulated in the actuality will be the indetermination of the certain degree because of the exterior and inside reason. The classic control theories based on the feedback primarily design the parameter of the controller according to the error, to reduce influence of the indetermination. But, the classic control theories can not give the valid method to design the controller with the analytical means. Modern control theories provide the analytical design means for controller, without considering the error of the model and the indetermination of the systems. The H control can make up blemish of modern control theories and classic control theories and deal with the indetermination of the controlled object effectively. It is very difficult to design the controller with traditional method, so the H controller can not be used widely.The genetic algorithm is a random global search algorithm, which imitating the mechanism of living creature evolving and the nature choosing. Its characteristics include: operating to the coded parameter; dealing with a series dots at the same time to protect local convergence; possessing characteristics for parallel computation to increase calculating speed.This paper presents a new method for designing robust H control system, in which the design objectives are converted into the inequality constraints and the selection of weighting functions in H controller design is expressed as a multi-objective optimization problem. The H controller, which satisfies given frequency-domain and time-domain performance indexes, can be found by using genetic algorithm (GA) to optimize the weighting functions.