Simulation on aircraft generator control system using genetic algorithms

In this paper, a comprehensive model of cascade brushless doubly-fed aeroplane generator is developed on the analysis of aeroplane generator´s structure and the principle of variable speed constant frequency electricity generating. Furthermore, a simulation model of brushless direct current machine derived from this model is established in SIMULINK development environment to verify the validity of the model. In order to overcome the defects of hypo-best PI parameters by conventional method of trial and error, the real-coded genetic algorithms is adopted to optimize the parameters of rotate speed in which the fitness function is considered sufficiently for capturing systemic error and controlled output. The simulation results show that the best PI parameters can be quickly educed by adopting stator field orientation vector control theory and genetic algorithm. It is guaranteed that the power system output remains constant frequency and constant voltage of three phase ac under the circumstances of speed changing. However, the control strategy is considered as valid and effective.