The Research On Causes And Solutions Of Nonideal Control

For the industrial process control,there may be model mismatch because of the change of the external environment.The influence of the model mismatch and the immeasurable disturbance will lead to the nonideal control.The nonideal control means that the system is in a nonideal state for various reasons.Of course,nonideal control is undesirable in process control.Thus,it is instructive and meaningful to research on the reason of non-ideal control and provide corresponding solutions.In terms of linear model predictive control,this paper provides two methods to solve the problem of nonideal control.The first one is to perform the errors feedback correction.By calculating the error between the model output and the plant output and take the proportion,derivative and integration of the error,an algorithm based on PID feedback correction is provided.The superiority of this algorithm is that it considers the error information at current time,as well as the error information at previous times.The feasibility of the proposed algorithm is illustrated by the examples of distillation operation and heavy oil fractionator.The other one is to analyze the statistical characteristics of the model parameters.The actual process model may changes because of external uncertain factors.In order to estimate the model parameters of the real system,a kalman filter correction method is proposed.In the algorithm,firstly,the parameters of the model are assumed to conform to certain distributions;secondly,the state of the system is estimated by the kalman filter algorithm;finally,the optimal estimated state is used in the MPC optimization problem to obtain the optimal control solution.Simulation results of two examples indicate the effectiveness of the proposed method.