| The reactor is widely used in chemical,pharmaceutical,food and other industries.It is a pressure vessel used to complete sulfurization,nitrification,polymerization,condensation and other process.T The degree of automation of reactor control at home and abroad is low.How to achieve accurate temperature control of reactor has become the bottleneck of manufacturing.The research on the online control technology of reactor has also attracted more and more attention.On the basis of fully investigating and analyzing the temperature control system of the reactor at home and abroad,considering the temperature changes of the processes of heating,pelletizing,pre-steaming and steamdissolving,and aiming at the time-varying,non-linear and large time-delay characteristics of the temperature dynamic characteristics of the existing jacketed reactor,this paper adopts the research method of combining theoretical analysis,simulation and experimental verification.By means of system identification,modern control,predictive control,computer simulation and other theoretical tools,the key problems of temperature control scheme,model identification,online temperature control and software development of temperature control for reactor are studied in depth.The main contents of this paper are as follows:(1)A set of temperature control scheme for the pelletizing process of reactor was designed.Single-Input Single-Output(SISO)model of pelletizing process control system was established with steam valve opening as control(input)and solution temperature in reactor as control(output).Based on the specific production process of pelletizing process,the temperature control requirements of pelletizing process are summarized and analyzed.According to the input and output historical data,the segmentation modeling and segmentation control mode of the temperature control system are confirmed.At the same time,the time-delay coefficients of three stages are selected,and the DMC control scheme based on subspace identification to obtain the temperature model is selected.(2)Identification of temperature model of reactor in different stages of pelletizing process has been completed.A subspace identification framework of deterministic system is constructed.Based on historical data,the N4 SID subspace identification algorithm is applied to identify the temperature model of reactor facing the pelletizing process,and the state space models of three temperature control stages,namely,heating stage,pelletizing pre-steaming stage and steam-dissolving stage,are obtained.Based on matlab,the fitting degree of historical temperature curve and model validation curve in each stage is 92.62%,88.92% and 93.90% respectively,which confirms the validity of the system model.(3)The temperature control of DMC in reactor was studied.The conversion from DMC algorithm of common object to DMC algorithm of time-delay object is completed,and the appropriate control parameters are selected.The temperature model of reactor identified by subspace is used as the predictive model of DMC controller,and the optimal temperature curve of pelletizing process obtained from the analysis of quality optimization system of energetic material production process is used as the reference trajectory.The temperature controller of reactor is designed in three stages.The tracking control of the reference trajectory is realized by Matlab,and the control simulation of each stage and the whole process is completed.(4)The development of control system software has been completed.For the temperature control of the reactor in the process of pelletizing,the software data transmission module of the online control system and the three interfaces of online monitoring,system identification and control simulation are developed by using Visual Studio MFC,and the DMC control algorithm is compiled to give it online control function. |
PDF Full Text Request