Research On Simplification Strategy Of Cascade Controlled Object Control Structure Of Bottom Chemical Process Under Predictive Control Architecture

In the chemical process,in order to solve the unfavorable characteristics of controlled objects such as constraints and multi-variables,it is usually necessary to introduce advanced process control represented by model predictive control on the basis of the conventional control structure at the bottom layer,and establish a system architecture of predictive control to achieve better control performance.However,when the underlying control loop adopts a cascade control structure,it will cause adverse effects such as increased complexity of the system architecture and prolonged dynamic response adjustment time.It is an effective way to further improve the performance of predictive control system by solving the problem that the underlying control loop cannot be directly cancelled,and exploring the simplified strategy of the underlying cascade control structure to reduce the system complexity and adjustment time.First of all,this paper takes a type of serial continuous stired tank reactor as the controlled object of a typical chemical process,and establishes its accurate mechanism model based on MATLAB software,and then introduces a predictive control loop on the basis of the cascade control structure at the bottom to make it run stably,and establishes predictive control system.At the same time,the adverse effects of time and space complexity increase and prolonged adjustment time caused by the underlying cascade control structure are analyzed in detail.Therefore,simplification strategies of canceling the main loop or secondary loop of the bottom cascade are proposed,and different control structures are derived by theoretical derivation.Then,the predictive controller is redesigned based on the conversion relationship of the predictive model,and the simplification of the underlying control structure is achieved through the transformation of the output manipulated variables of the predictive control system.Secondly,in order to simplify the underlying control structure to the greatest extent,a simplification strategy is proposed to cancel the primary and secondary loops of the underlying cascade at the same time.By studying the adjustment strategy of the predictive model,the predictive model and the MPC controller are re-obtained.At the same time,the primary and secondary variables in the underlying cascade loop are unified into the predictive control controlled-variable and fed back to the predictive controller,so that all the predictive control system output controlled variables are converted into given values of the valve opening,which simplifies the underlying control structure.The final simulation results showed that the above three control structure simplification strategies can effectively reduce the system complexity and dynamic response adjustment time,and also improve the overall performance of the control system.In the case of canceling the bottom cascade loop alone,the simplified strategy of canceling the flow sub-loop is more effective.When the bottom-level cascade main and sub-loops are canceled at the same time,the system complexity and adjustment time can be reduced to the greatest extent.