Design Optimization And Applycation Of Active Desturbance Rejection Controller

Active Disturbance Rejection Controller (ADRC) is a kind of nonlinear controllerwhose design is independent on the model of the controlled plant. As the maincomponent of ADRC, the nonlinear Extended State Observer (ESO) can observe andcompensate the uncertain time-varying disturbance, which makes ADRC have strongerrobustness and better disturbance rejection performance. However, ADRC has too manyparameters whose tunings are very complicated, and its stability cannot be guaranteedeasily. The unclear relation between the control performance and parameters of ADRClimits its application. In order to improve the performance of ADRC, this paper analyzesall the components of ADRC deeply and proposes an integrated optimization designmethod.First, the third-order nonlinear Extended State Observer (ESO) is analyzed as themain object. The Lyapunov function with special form is constructed on account of thenonlinear function used in ADRC. And a sufficient condition for the asymptotic stabilityof the third-order ESO is derived and then is verified by simulation. Furthermore, thefrozen-coefficient method and pole assignment method are employed to extend thebandwidth of disturbance observation. The variation of the assigned poles with thenonlinear coefficient is analyzed. Consequently, an optimized parameter configurationmethod is proposed to minimize the variation of the disturbance observation bandwidthwith observer state. The simulation comparison is conducted to show the effectivenessand superiority of the method.Second, based on the analysis of the traditional nonlinear feedback function used inADRC, a modified nonlinear function is constructed for improving the controlperformance. The stability of the improved nonlinear PD controller is analyzed and anew optimal design method is proposed using pole assignment. Consequently, theintegrated optimization design method of ADRC is summarized and the specific designsteps are provided.At last, the application of the design optimization method in servo system isstudied. The mathematical model of Permanent Magnet Synchronous Motor (PMSM) isbuilt, and its disturbance characteristics are analyzed. Then, the integrated optimizationmethod is applied to design the ESO and nonlinear PD controller to meet theperformance requirements. The experiment results show that the proposed method ismore superior than the traditional one.