Optimal Setting Control For Two-Layer Structure Systems

External disturbances exist widely in various systems.To guarantee stable and desired controlled outputs under complicated and changeable operating environments is one of the difficulties in control science.Some advanced control methods can solve this group of problems partly.But due to the constraints of technology and economy,the conventional feedback control methods,such as PID,are still widely used in a considerable part of practical engineering.It will lead sizable economic losses to change the structures or parameters of the basic controllers to adapt external disturbances and operating environments.This thesis considers a group of two-layer optimal setting control strategies with more convenient operations and less cost.The upper layer sets and adjusts the target values and the set values,while the under layer carries out the tracking control of the target values.The fundamental of this kind of methods is to adjust the set values of control systems online,so as to control the systems effectively under external disturbances and changeable operating environments without changing the basic controllers or stopping the production devices.The basic theoretical framework of the proposed methods is provided in this thesis,and also Matlab numerical simulations are carried out.The main research work is listed as follows:1.For a group of linear systems subject to deterministic external disturbances,a two-layer optimal setting control strategy based on Lyapunov stability theory and linear matrix inequality(LMI)approach is considered.Firstly,the tracking controller is designed for the ideal models without any external disturbances.Then the system subject to external disturbances is considered.The set values are adjusted to drive the controlled outputs to track the target values and thus guarantee the performance function to be optimized within a certain range.2.For a group of linear systems in which the states can not be measured directly,an observer based optimal setting control strategy is designed.The dynamic compensators are designed for the ideal models.When there are external disturbances,the set values are adjusted in the light of the observers to make the controllers designed for the ideal systems can be applied to the systems with external disturbances.3.For a group of servo controlled systems subject to deterministic external disturbances,an observer and PID controller based optimal setting control method is designed.A numerical simulation of manipulator path tracking control is given to demonstrate the effectiveness of the proposed method.4.For nonlinear systems subject to non-Gaussian stochastic external disturbances,a two-layer optimal setting control method based on the entropy optimization principle is designed.By adjusting the set values on line,the uncertainty of the performance deteriorations caused by external disturbances is minimized and the expectation of the deterioration is drawn close to 0.The optimal setting control problem is transferred to a multi-objective optimization problem by combining the indexes of the entropy and the expectation,and a new optimization method named Pareto-DE is proposed to solve the problem.