Research On Disturbance Observer Based Model Predictive Control

The time-delay,disturbance,noise and other factors in industrial process control will affect the control performance and the economic benefits of the enterprise.The MPC method can effectively deal with the disturbance of multivariable system,but the effect is not good,and it cannot deal with the time-delay and disturbance at the same time,so it needs to be improved to better deal with the time-delay and disturbance.The main work is as follows:Firstly,for the multi constraint system with external disturbance and multivariable coupling disturbance,the existing methods usually are usually based on feedback and feedforward compensation,which cannot guarantee the optimal output of the system.Therefore,a disturbance observer based multivariable non-minimum state space model predictive control method is proposed in this paper.The disturbance observer is used to estimate the total disturbance(external disturbance and internal disturbance),and the estimated value and output variable are introduced into the state variable at the same time and fed back to the front-end model predictive controller for rolling optimization,which avoids the design of the state observer,and then solves the multi constraint problem by using the quadratic programming method,and ensures the optimal output.Finally,the effectiveness of the proposed method is verified by simulation in the gas fractionation unit.Secondly,for a system with time-delay,external disturbance and noise,the existing control methods cannot deal with the problem of time-delay and disturbance at the same time.Therefore,a double disturbance observer based model predictive control method is proposed in this paper.This method transforms the time-delay into a kind of disturbance momentum,estimates itwith the first disturbance observer and feeds it back to the front-end model predictive controller for compensation.Then estimates the external disturbance and noise with the second disturbance observer and compensates with the controller output,which can deal with the time-delay and disturbance problems at the same time and improve the control performance of the system.Finally,the effectiveness of this method is verified by simulation in the control system of BLDCM.