Design of adaptive Fuzzy-Smith controller based on improved genetic algorithm

For the complex controlled object with time-delay and time-varying, this paper presents an adaptive Fuzzy-Smith control system, combining the fuzzy logic controller optimized by Genetic algorithms with the Smith Predictor. The least-square parameter online identification technology is adopted to accomplish the self-adjusting of Smith Predictor according to the parameters variation of the object model. Meanwhile, the improved genetic algorithm adjusts the fuzzy logic controller to adapt to the object with parameters variation. The proposed method is applied to an aircraft deicing vehicle under MATLAB, and the result of the simulation illustrates that the system has good dynamic performance and robustness.