Active Disturbance Rejection Control Theory And Applications In Cold Strip Rolling Mill

There are multiple independent and correlative control systems in cold rolling production process, which are complicated nonlinear systems with large uncertainties. Therefore, the control of uncertain nonlinear systems is of great important to realize the effective control of cold rolling production process. Nonlinear systems are complex and diverse. Different control methods have been proposed for different nonlinear systems. However, these control methods usually base on mathematical models, which limit these algorithms to apply in practice. As a practical nonlinear control method, active disturbance rejection control technology does not rely on internal mechanism and external disturbance model of the controlled system, because uncertain nonlinear can be estimated and compensated in real time. Active disturbance rejection controller is feasible and easy to be applied in practice, and has been widely used in many fields. In this dissertation, the problems of active disturbance rejection control for uncertain nonlinear systems are studied. The stability of nonlinear active disturbance rejection control system is analysed through combining singular perturbation theory and active disturbance rejection control technology is applied in cold rolling production process. The main research results are stated as follows.Firstly, the stability of nonlinear active disturbance rejection control for uncertain nonlinear system which can be simplified as integral series system is analysed by constructing singular perturbation system. The stability analysis and theoretical proof of nonlinear active disturbance rejection control for SISO and MIMO systems are given.Secondly, active disturbance rejection control for unified chaotic system is proposed. Sufficient conditions for asymptotic stabiliby of the closed-loop systems are derived and stabiliby of the closed-loop systems is analysed. The simulation results confirm that active disturbance rejection control strategy can suppress the chaotic phenomena. At the same time, the work can be regarded as the preliminary work to suppress the chaotic phenomena during rolling using active disturbance rejection control.Thirdly, extended state observer that switches between two gain values is proposed for uncertain nonlinear system with output noise. The convergence in the presence of output noise is analysed and illustrates the tradeoff when selecting the observer gain between state reconstruction speed and robustness to uncertainties on the one hand versus amplification of noise on the other. Based on this tradeoff, extanded state observer with switched-gain is designed. This scheme is able to quickly estimate states of the system during large estimation error with high gain values and reduce the negative influence of output noise during in a neighborhood of the origin of the estimation error with low gain values. The convergence of extended state observer with switched-gain is analysed.What’s more, active disturbance rejection control techonology is applied in hydraulic servo system of rolling mill. Based on active disturbance rejection control, the internal dynamic and external disturbance can be estimatied and compensated in real time, and the both sides of hydraulic servo system can be quickly synchronized. The frequency domain anaylsis, simulation results and filed experiment results demonstrate the effect of active disturbance rejection control.Finally, active disturbance rejection control is employed in the strip gauge control of cold rolling with roll eccentricity and vertical vibration. Active disturbance rejection repetitive control of GM-AGC system is proposed. The roll eccentricity can be compensated in real time to reduce fluctuation of strip thickness without the eccentricity signalcharacteristic, such as amplitude and phase. In addation, the strip gauge control system model in condition of vertical vibration is built by coupling with hydraulic servo system, mill stand structure system and rolling process system. Then, active disturbance rejection cascade controll for strip gauge control system based on this model is designed. Simulation results show that the proposed control strategy can improve robustness of the strip gauge system.