Perfotmance Analysis And Impronement Of Linear Adrc Based On Total Disturbance

Active disturbance rejection control(ADRC)takes the series integral as the normative model of the system.The information outside the normative model is regarded as the total disturbance of the system.The total disturbance is estimated and compensated in real time by the extended state observer(ESO).Therefore,ADRC has strong robustness and anti-interference performance and does not depend on the exact model of the system.The number of controller parameters and the difficulty of adjustment is reduced by the linear ADRC,which facilitates theoretical analysis and engineering applications.The realization of the superior performance of ADRC lies in the innovative introduction of the system.Therefore,it is necessary to analyze the influence of total disturbance on the performance of ADRC.The research contents of this paper include the following aspects:First,The influence of total disturbance on the linear ESO estimation ability is analyzed.The transfer function from the total disturbance of the system to the estimation error of the linear ESO is obtained,so that the linear ESO estimation error has a unified metric.The "horizontal contrast" between linear ESO state estimation errors is achieved for the first time.The relationship between state estimation error and observer gain,amplitude,maximum value and stability time of each state estimation error are analyzed and compared by frequency domain and time domain method respectively.The relationship between state estimation error and observer gain,the amplitude,the maximum value and stability time of each state estimation error are analyzed and compared by frequency domain or time domain method.Under the more relaxed assumptions,this paper proves that the improved linear ESO,which makes full use of the known information of the total disturbance,has higher observation accuracy.Second,The influence of total disturbance on the performance of the linear ADRC closed-loop system is analyzed.Under the premise of ignoring the ESO's estimation error,the controlled system becomes a series integral model after linear ADRC compensates for the total disturbance.And the ideal closed-loop model is obtained through pole configuration.The effect of linear ESO estimation error on the closed-loop system is analyzed,and it is found that the closed-loop output of ADRC is actually composed of the target transfer function(the reference signal as the input signal)and the error transfer function(the total disturbance as the input signal).In this paper,the influence of total disturbance on the closed-loop output of the system is qualitatively analyzed,and it is pointed out that there is a constraint relationship between the linear ESO parameters and the control law parameters.Third,the linear ADRC is improved based on the estimation error compensation.Considering the influence of linear ESO estimation error on the linear ADRC closed-loop system,an improved method is proposed to compensate the estimation error in the error feedback control law.The influence of the reference command signal on the control performance is also analyzed,and the high-order differential of the feedforward compensation reference value can effectively reduce the amplitude of the control signal.Finally,the stability of the ADRC improvement scheme based on estimation error compensation is proved.Fourth,the proposed method of this paper is simulated and verified in the cascade control system.Through the simulation of the superheated steam temperature system with long time delay and large inertia,the proposed method has strong robustness and better control effect when the dynamic characteristics of the system vary with the operating conditions.The simulation results of the mechanical automatic transmission displacement cascade control system show that the proposed method can effectively achieve fast tracking and has stronger anti-interference ability.