Optimization And Control Of Extractive Distillation For Azeotropic Formic Acid And Water Separation

Formic acid is an important chemical raw material,which is widely used in chemical production.Formic acid is produced industrially by hydrolysis of methyl formate,and the resulting formic acid product contains unreacted water.However,formic acid and water form the maximum-boiling azeotropes,which cannot be separated by conventional distillation.For the separation of formic acid and water,the BASF process uses liquid-liquid extraction and distillation to dehydrate the formic acid;and the Kemira-Leonard process purifies the formic acid through thermal integration pressure-swing distillation.In this paper,the separation process of formic acid and water was studied,and sulfolane was selected as the entrainer to design an extractive distillation process,which was optimized using a genetic algorithm（GA）with the minimum total annual cost（TAC）as the optimization objective.To further integrate and enhance the process,the extractive dividing wall distillation process was designed to reduce energy consumption and reduce capital costs.The optimization results were compared with the optimal extraction and distillation process and the thermal integration pressure swing distillation process using economic indicators of TAC and environmental indicators of CO₂ emissions.Since sulfolane is thermally decomposed at high temperatures,the entrainer recovery column is operated under vacuum conditions to prevent the thermal decomposition of sulfolane.This results in a mismatch of pressure conditions between the extraction distillation column and the entrainer recovery column during process integration and intensification.The extractive dividing wall distillation column needs to be operated under vacuum conditions,which leads to a significant increase in the amount of sulfolane,thus the extractive distillation process has a lower TAC and less CO₂ emissions.The subsequent Proportional Integral（PI）control and the advanced Model Predictive Control（MPC）were studied using the process.The control performance of the temperature control structure with PI control and MPC was compared by introducing feed flow disturbances and composition disturbances to determine the dynamic response,respectively.By directly observing the oscillation,overshoot and setting time of the dynamic response,it can be seen that the MPC can significantly improve the control performance of the variables that are poorly controlled by the PI control.For the dynamic response curves that cannot be compared visually,the error squared integral（ISE）of the control performance index is used to evaluate the dynamic control performance.Although the ISE value of the MPC of the extractive distillation column is slightly larger than that of the PI control,the overall control performance of the MPC is better than that of the PI control in temperature control,and for the control of product purity,the control performance of the PI control for composition disturbance is better than that of the MPC,but the overall control performance for product purity is similar.To further improve the control performance by overcoming the effects of pressure variations and rejection of composition disturbance,the temperature difference control structure was used in both PI control and MPC,and its dynamic control performance was tested in the same way.The control performance of MPC was significantly better than that of PI control in terms of temperature difference control loop and product purity,further confirming the superiority of MPC for the control of complex distillation processes.