Intelligent Optimization Control For Aeroengine

Aeroengines are complicated, non-linear and multivariable control systems, the models of aeroengines are under conditions of uncertainty and time varying over full flight envelope. PID controllers are still the most popular approach to control aeroengines, due to their simple structure and easy of adjusting the parameters. The parameters of the PID controllers, however, are very much dependent on the accuracy of the models. This paper studies the control algorithms for aeroengines based on intelligent optimization. As a new intelligence system, artificial immune system is a new method to deal with difficult control problems. Based on immune feedback mechanism of biological immune system, this paper presents an optimization control system, which combines immune controller and PID controller. Genetic algorithms have been used to search the group of PID controller parameters in global range independent on its characteristics.It provides a new method for tuning of the PID controller parameters. Two methods are proposed using genetic algorithms and neural networks and applied to optimize the aeroengine control systems in the full envelope. First, genetic algorithm is used to initialize the weight value of the neural network in order to accelerate its learning convergence; Then, genetic algorithm is used with a new indirect binary matrix codification to design the neural architecture.This control system can design the neural topoligy and train the weight of the neural network adaptively for the change of external environment.