| Distillation is one of the most important and applied widely separation technologies.The DWC can greatly reduce the equipment investment and energy requirements for the distillation process in chemical production.In terms of theoretical research,the optimization of equipment parameters and the selection of control strategies have been focused of experts and scholars at home or abroad.After reviewing most references,we found that DWC is a device integration process,which have highly complex mechanical structures and process parameters.It is regarded as a complex controlled system with nonlinear processes and time delays.In addition,the strong interaction between input and output multivariate correlations makes DWC more difficult to control than traditional distillation.Obviously,PID is not the best selection for DWC.Based on the summary of past research,this paper propose that use the sliding mode control(SMC)to control DWC,which can lay a theoretical foundation for the application of SMC in complex equipment in the process industry.The optimization of the steady state parameters and the control of dynamic process are studied as follows:(1)The "four-column model" of DWC was simulated and calculated to build a steady-state foundation for the separation of three-component mixtures.The sensitivity analysis was carried out with the aim of separation efficiency and minimum heat duty.Through analysis and comparison,more appropriate process parameters were obtained.Based on the sensitivity analysis,the column was optimized globally for response surface optimization and particle swarm optimization.(2)Based on the optimized steady-state structure,the dynamic control study of the three-component mixture separation process in the "four-column model" was conducted.We built temperature control(TC)and temperature difference control(TDC)for the system.By adding the disturbance of the feed flow and comparing the control effects of them,we found that both control structures can make the variables return to stability in a short time.However,compared with the TC,the TDC has improved significantly the control performance of the system.Main performance: decrease of maximum overshoot and decrease of steady state margin.In particular,the output of intermediate components in the column can be controlled stably.Then,we got the response curve of the process and derived the input-output data relationship.(3)According to the obtained data relationship,MATLAB software was used to mathematically model the dynamic control process of DWC with transfer function as the model,and then obtain the transfer function matrix between variables.Using the principle of RGA to carry out variable screening,we got the best match that can reflect the relationship between the manipulated variable,the disturbance variable and the controlled variable.(4)The SMC and PID were constructed respectively based on Simulink and Smith lag compensation link was introduced.Under the disturbance of the feed flow,the step response curve of the system was obtained and compared. |
