Research On Predictive Control Based On Intelligent Algorithms

Along with advancement of industrial control demand, development of control theory and computer technology, a predictive control algorism is produced with effective control and strong robustness, which is applicable to complex industrial processes and the control system, and is successfully applied in petroleum, chemical industry, metallurgy and mechanism. Studying of predictive control based on intellective algorithm has important theoretical and realistic signification.This text introduces the basic theories and structures of the predictive control first, discussing the predictive model of DMC, the feedback and its stability and robustness. Connected down to introduce the basic theory of neural network modeling detailed. It also introduces the genetic algorithm and differential evolution algorithm's basic theories and methods and advances a method to improve differential evolution algorithm, meanwhile using it to optimize the parameters of the neural network model.Based on the theories above, the text brings forward a predictive control method based on intellective algorithms which use neural network to establish system model and use the genetic algorithm and differential evolution algorithm to do rolling optimization. The text makes the predictive control simulations basing on the lag nonlinear model, restriction nonlinear model and the changing parameters nonlinear model respectively. The result shows its application. This text also compares different parameters for the result of the predictive control. The result shows that different parameters have big effects for the control's speediness, veracity and stability. At last, the text respectively uses GA and MDE to do predictive simulation to realize the rolling optimization and compare the two results. The two simulation results show the rolling optimization of predictive control based on MDE is better than based on GA on the whole.