Satisfactory Optimization Control Based On Fuzzy Goals And Fuzzy Constraints

It is a great development from regularization-based control to optimization based control in the industry process control, thus all kinds of control requests can be coped with performance criterions. The traditional optimization technique usually combined the fundamental goal of the controller and performance criterion into a single performance index to be minimized by applying technique based on Bellman's principle of optimality or Pontryagin's minimum principle. This view of control is designed to obtain the best solution. But as the industry becoming more complex and larger, a constrained multi-objective and multi-freedom optimization control problem (CMMO) has been proposed in dynamic uncertain environment, which makes it difficult to achieve the best solutions. So It's necessary to find a satisfactory solution rather than a best one. What's main research of the thesis is to make some research, based on the basic thought of the satisfactory control, into some problems of satisfactory optimization control with fuzzy objective and fuzzy constrains in the complex industry process. The whole thesis includes five chapters and the main contents and conclusions are summarized as follows:In chapter one, first of all, we review the main characteristics and challenges of each stage of the whole development procedure of control theory. Secondly, make a introduction of the background, fundamental conceptions and structures of the satisfactory control, and summarized recent research and development of it. Finally, point out the advantage of the satisfactory control which oriented application based on predictive control.In chapter two, firstly we make a summary of fuzzy set and fuzzy number, and introduce a kind of fuzzy optimization named fuzzy object-resource. Secondly make a introduce of a typical multi-objective optimization.In chapter three, firstly introduce the fundamental conceptions and methods of generalized predictive control (GPC). Secondly, we give the mathematic fuzzy environment and satisfactory decision-making. Then make a summary of the method how to resolve a satisfying optimization problem, finally, we research a kind of method based on fuzzy programming to solve a satisfied optimization problem with some constrains and this method has been verified by the simulation results.In chapter four, problem about parameters of GPC on-line tuning in frame ofsatisfactory control is researched, and a satisfactory model for multi-parameters on-line tuning is defined, and proposed a formula based on grade conception and one-dimension searching method. It's shown that the system's performance can be improved by this approach.In the final chapter, the conclusions of the thesis and future research directions are presented. A lot of researches are deserved further efforts to make up satisfactory control theory, which will contribute and enrich the development of modem complex industry process control.