Font Size: a A A

Design Of Controllers With Optimized Structure And Its Application In Maglev Ball Control System

Posted on:2017-11-03Degree:MasterType:Thesis
Country:ChinaCandidate:Y Q ZhangFull Text:PDF
GTID:2348330566456667Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Automatic control systems have been widely used in modern industry,defense industry and other fields.The stability of control systems is mainly dependent on the regulation of the controller.So far we have an enormous spread of control methods,including PID control,fuzzy control,adaptive control,auto disturbance rejection control,etc.The PID controller is widely used in process control due to its simple structure,adjustable parameters and strong robustness.However the structure of PID controller is fixed,so controllers can only adjust the parameters,due to which it is just used for linear systems.PID controller often fails to control complex and nonlinear control systems.Design of controllers includes structure determination and parameter adjustment.Parameter adjustment needs to be carried out under the premise of structure determination.At present,the research on structure optimization of controllers is scarce,and parameter optimization is mainly based on the traditional control theories or intelligent optimization algorithms.Due to the limitation of controllers' structure,the performance of controllers can not get a breakthrough improvement.The development of controllers needs to break the constraint of design experience and realize the automatic design of structure.Intelligent optimization algorithms are inspired by the law of nature to solve complicated optimization problems.They have high universality,being suitable for parallel computing and global optimization.Compared with the traditional optimization algorithms,intelligent optimization algorithms are more and more applied to optimization design of control systems,which can effectively improve the performance of control systems.Intelligent optimization provides an effective way for the automatic design of controller structure and parameters.In order to realize the automatic design of controllers,this paper proposes library-based design of controller with optimized structure."Library" is a set of controller design elements,including the component library,the performance index library,the optimization algorithm library and the constraint library.Component library includes the basic unit of controller structure.Performance index library includes various indicators of performance evaluation.Optimization algorithms library includes a variety of search and optimization algorithms for structure and parameter.Constraint library includes limiting factors in actual control system.In this paper,controller design process includes 4 steps.The first step is to select a plurality of basic units from the library as a component of controller structure,for subsequent optimization of controller's structure.The second step is to analyze control systems,consider the constraints of actual systems,and establish control system simulation environment.The third step is to select objective function in the performance index library.The fourth step is to select a certain algorithm in the optimization algorithm library to optimize the structure and parameters,apply the controller designed by simulation to actual control systems,and verify the performance of systems.PID controller is composed of three typical parts,which are proportion,integral and differential component.In this paper,a method for optimizing structure is presented.The control component can be connected in series,parallel and feedback connections and other forms,and each components contains a parameter.Intelligent optimization algorithms are used to optimize the structure and parameters of controllers.In this paper,the matrix representation of the controller structure is presented,and the connections between components are expressed by a matrix.The optimization of structure is embodied in component relation matrix under given design index.Simulation experiments are carried out on several typical systems(non minimum phase system,high-order systems and nonlinear systems)with different performance indexes.Compared with the existing control methods(2-DOF PID controller,sliding mode controller,etc),controller with optimized structure is better.In addition,the controller in this paper is also used practically in the maglev ball control system.The controller with optimized structure can realize the fast and steady control of the maglev ball control system,which demonstrates the feasiblity and superiority of the proposed method in nonlinear system control.
Keywords/Search Tags:Optimal Design of Controller, Structural Optimization, Library-based Controller Design, Intelligent Optimization, Differential Evolution
PDF Full Text Request
Related items