Research And Simulation On Global Predictive Control Algorithm And Wiener Model

Predictive control algorithm appeared in the mid-70s, after near 30 years'development, not only has made great progress in theoretically, but also obtained good results in the practical application, its formidable vitality has received enormous attention by control field, it already became the model of the process control.The generalized predictive control is one new computer control method produced in 80s, is one of the most representative algorithms in the predictive control. The algorithm has received attention by the domestic and foreign control theorists and the industrial world as soon as it appears, and became the most active kind of predictive control algorithm in the research area. But because its big calculation amount, especially in the many-variable system, and can not give consideration to both rapidity and over modulation, so the algorithm does not suit for system which is strong real time or request request high rapidity. Therefore improve generalized predictive control algorithm becomes the hot spot topic.The main research content of this article are as follows:(1) Aiming at solving the problem of big calculation amount of generalized predictive control (GPC) algorithm itself and can not give consideration to both rapidity and over modulation, proposed one kind of improved generalized predictive control algorithm.(2) In view of the fact that the track speed of generalized predictive control track speed is slow, introduced a reference trajectory self calibration generalized predictive control strategy.(3) Proposed one kind of nonlinear generalized predictive control algorithm based on the wiener model. Mainly studied a strategy based on two step trundle optimization, this strategy decomposed the nonlinear forecasting problem to dynamic optimized question for the linear model and the static state root question of nonlinear model, Proposed one new approximate method to inferior equation which cannot obtain the real roots occasionally.The article obtains the achievement are as follows:(1) The fast generalized predictive control algorithm had reduced the algorithm calculation amount greatly, can give consideration to both rapidity and over modulation, make the very good theory preparation for the following research scholar.(2) The reference trajectory self calibration generalized predictive control strategy can make the system response walk along to the expectation path we planed, and arrive at the setting value in assigned time.(3) The new approximate method proposed solved the question of the occasionally inferior equation which not able to obtain the real roots, this method already could close strive for the control effect of directly method.