Research On Nonlinear System Control Problem Based On Characteristic Model

To address the difficulties of establishing accurate dynamic models for physical ob?jects and the challenges to complex system control problems,academician Wu Hongxin proposed the characteristic model adaptive control method based on the all-coefficient adaptive control theory.This method considers not only the controlled plant,but also the control performance requirement.Since that the characteristic model has been widely used in aluminum electrolysis process,satellite attitude control systems,hypersonic ve-hicle control systems and so on,the researchers gradually explore the control theory of theoretical level.However,considering the definition of characteristic model,we should study the modelling method first.Then a controller can be presented,which can guarantee stability of the closed loop system.The framework of the control method has been formed by the above process,which lays the foundation of practical application.The modelling method,controller design and stability analysis are studied in this book.The main contents are summarized as follows.Firstly,we study the characteristic modelling method of nonlinear systems.Re-searchers focused mainly on the effectiveness of the control method,while the impor-tance of theoretical basis of the method is ignored.The stability was just analyzed the closed loop of characteristic model and the controller.In order to perfect the control theory,we should deal with the modelling method.We propose the corresponding the modelling method of the nonlinear systems with different assumptions,present the pa-rameters range,and quantify the modelling error at the sampling instants.The above work provides a basis of the consequent analysis.Secondly,the stability of the closed loop system is analyzed.Considering that the characteristic model is a kind of sampled data based control method,the modelling errors are equivalent to the consistency conditions.The theoretical derivation of the modelling method leads to new consistency conditions.Therefore,we study the exact discrete time closed loop system by studying the stability property of the approximate discrete time closed loop system.The controller of linear structure is designed based on the estimation of the characteristic parameters,and an adjustable parameter is intro-duced to improve the amplitude of the controller.The above logical line provides the framework of theoretical analysis for characteristic model.Then,the distributed control problems based on the characteristic model for com-plex nonlinear system are studied.The control method of the first-order characteristic model is simulated in the cryogenic transonic wind tunnel.We have the similar research conclusions in multiple-input multiple-output case,while it also has its particularities.This thesis studies the modelling of complex systems with coupling MIMO subsystems,and analyzes the stability of the closed loop system.We separately derive the character-istic model and design the controllers on the subsystems in wind tunnel.We also prove the stability of closed loop system.At last,the effectiveness of the control scheme based on the MIMO nonlinear system distributed characteristic model method is illustrated by a simulation of wind tunnel under certain conditions.Finally,we study the control problem of nonlinear systems based on the abstract models.In order to instantiated the abstract model,we design the controller with a kind of simple systems being the abstract model.The output interface dynamics is structured with the output signal of original systems and the input signal of the abstract model,and the input-to-state stability of the closed loop system is proved with the controller.At last,the differences and connections between the abstract model and the characteristic model are given and the emphases the two models are compared.