Process Of Separating Light Hydrocarbon With Mid-heat Pump Assisted Dividing Wall Column And Its Operation Optimization Research

C₁-C₄ light hydrocarbons are widely existed in ethylene and petroleum exploration,petroleum processing,alkane dehydrogenation and coal retorting industries.The separation of ethylene,propylene and other monomeric olefins is the foundation for the development of the polyolefin industry.However,the connentional rectification with high pressure,low temperature and double deethane column separation process suffers from extremely high energy consumption due to the the low and approximate boiling points of the diverse light hydrocarbon isomer mixtures.Therefore,a light hydrocarbon separation process of the middle heat pump dividing wall column（MHP-DWC）by integrating internal thermal coupling and external thermal coupling has been developed in this paper,while the operation optimization research has also been conducted.Internal thermal coupling,that is,the de-propanizer and the de-ethanizer stripping section are combined into a low-pressure DWC with a common stripping section;The external thermal coupling is MHP,where DWC is equipped with an intermediate reboil,and the heat source is the overhead gas that is compressed and heated by the compressor in the pre-fractionation part of DWC.The whole process simulation of a light hydrocarbon separation unit with 92.3 t/h feed shows that the improved process can ensure the product yield and quality of the system,and save energy costs of 7.5774 minlions yuan per year co MPared with the existing process.Operation optimization is based on the process simulation data of PRO/II.Firstly,the Lasso regression function in the Python software calculation platform is called to calculate the influence weight of each adjustable operating variable on the given target to determine the decision variable;Secondly,we establish a PRO/II,Excel Software integration platform,use PRO/II COM as the interface between PRO/II and Excel,program and control the operation of PRO/II in Excel VBA,write and read simulation data code and optimize algorithms code such as genetic algorithm（GA）to control the optimization calculation process automatically.The calculation of the improved case shows that the optimal value of the five decision variables is Xopt=（63.9,1.67,3.99,2.99,0.51）when taking minimum exergy loss of the system as objective function.Co MPared with the operation before optimization,the total process exergy loss was reduced by 15.2%,saving energy cost 3.243 millions yuan per year.Moreover,the optimal value of the five decision variables is Xopt=（67.85,1.63,4.48,3.13,0.46）when taking exergy consumption of the system as objective function.The total exergy consumption was reduced by 15.8%,saving energy cost 3.173 millions yuan per year.Because the optimization process is driven by simulation data,our multi-variable,single-objective optimization method based on process simulation data and PRO/II,EXCEL,VBA（GA）software integration has universal applicability and is suitable for the operation optimization of all simulating chemical units that do not change the process structure.