| Taken design and optimization of an atmosphere-vacuum distillation in a refinery as engineering background of this dissertation, MESH (Mass balances, Equilibrium relations, mole fraction Summations and Heat balances) model of the large scale atmosphere-vacuum distillation equipment is established. Based on the flow simulation, taken feed locations of topping distillation column, atmosphere distillation column, and vacuum distillation column as optimization decision variables, and maximum annual net profit as objective function, the optimization is finished. A FORTRAN code is programmed to solve the optimization problem on the PRO/II platform with the second development. Good optimization results are implemented. Basically data of hot and cold streams that transfer heat with each other in the atmosphere-vacuum distillation process are extracted. Pinch analysis is carried out for heat exchangers network of atmosphere-vacuum distillation. Composite curves, grand composite curves, and duty contribution of hot and cold stream in different temperature intervals are obtained. MINLP model of heat exchangers network is founded with mathematical programming method. Branch and bound method is applied to solve the MINLP model. Heat exchangers network with splitting stream and without splitting stream are achieved. Advantage and disadvantage are compared for both heat exchangers networks. Suggestions are made how to select which of heat exchangers networks. Detail researches are summarized as below.1. According to craft of atmosphere-vacuum distillation, equations for one stage known as MESH equations, namely, overall material balance equation, material balance equation for each component, vapor and liquid equilibrium relationship for each component, stoichiometric constraint, and energy balance equation are formulated. SPR equation for state equation, API relationship for calculation of liquid density, PRETRO method for compute of transfer property, Glitsch approach for calculation of pressure drop of float valve tray, and simple heat exchanger model for calculation of heat transfer are selected in flow simulation. Flow rate, temperature, and pressure of each stream are obtained by solution of the model. Profile of temperature, vapor and liquid rate versus tray numbers reflect the distillation process and influence of feeds, products, and cooling cycle in the middle of column on the distillation process very well. The simulated results show good agreement with spot technological requirement. Creditable results of flow simulations provide flow optimization with firm foundation.2. Based on the flow simulation, taken tray numbers, flow rate of crude YY, components of crude YY, enthalpy, parameter of products, parameter of cooling cycle in the middle part of column, and parameter of side column as constraint conditions and taken all parameter being greater zero as inequality constraint conditions, taken feed locations of topping distillation column, atmosphere distillation column, and vacuum distillation column as optimization decision variables, taken maximum annual net profit as objective function, optimization model is established. The model is solved by the second development of FORTRAN program on PRO/II platform. Expected optimization result is achieved. Comparison of profile of temperature, vapor and liquid flow rate versus tray numbers in topping distillation column, atmosphere distillation column, and vacuum distillation column between flow with optimization and without optimization is done. The results show that profiles correctly reflect influence of change of feed location on the distillation process.3. The results with optimization and without optimization show that temperature and pressure have not great change which means the optimization flow satisfy process requirement. The major change is changes of product flow rate. According to price of products, the flow rates of higher price products increase greater while the flow rates of lower price products decrease. Although decrease of pump work contribute to increase of annual net profit, general profit of products makes most contribution to increase of annual net profit. The increase of annual net profit cost increase of heating duty and cooling duty. Annual net profit of optimization flow is higher 40.7 million RMB than that of flow without optimization.4. Data of hot and cold stream extracted from the optimization flow together other spot hot and cold stream construct heat exchanger network. Pinch analysis for heat exchangers network is done. Duty distributions of hot and cold stream in different temperature intervals are obtained. Composite curves and grand composite curves are plotted. For hot stream, hot end provides more heat duty than cold end. JNZY contributes most to the whole heat duty among all hot streams. For cold stream, the major streams needed heat duties are BTY and YY, and are mostly in the middle and bottom temperature intervals. The composite curves and grand composite curves show that smaller HRAT is, larger recovery heat is, smaller heating and cooling utilities are supplied, smaller capital cost of utility and annual operating cost of hot and cold utility is, but larger area of heat transfer is needed due to the increase of HRAT which result in increase of capital cost of heat exchangers. It shows that the capital cost of heat exchangers should be considered besides quantity of heat recovery. Pinch of heat exchangers network changes for different HRAT. 5. Applied mathematical programming, MINLP model for heat exchangers network including all of hot and cold stream is established. The model is solved by branch and bound method. Two types of heat exchangers network with splitting stream and without splitting stream are obtained respectively. For heat exchangers network without splitting stream, inlet and outlet temperature, duty, and area of heat transfer of heat exchanger, heater, and cooler are presented. Network without splitting stream has 142 heat exchangers, 8 coolers, and 8 heaters. For heat exchangers network with splitting stream, inlet and outlet temperature, flow fraction, duty, and area of heat transfer of heat exchanger, heater, and cooler are also presented. Network with splitting stream has 34 heat exchangers, 8 coolers, 4 heaters, 11 splitters, and 11 mixers. Grid plots for network with splitting stream and without splitting stream are obtained respectively. Grid plots could make clear situation of heat transfer for both hot stream and cold stream including inlet and outlet temperature, flow rate, duty, area of heat transfer, and flow fraction. Compared both types of network, total duties of both networks are same, capital cost and annual operating cost of heating utility and cooling utility in both networks are also close, but total area of heat transfer in network with splitting stream is greater than that of heat transfer in network without splitting stream. More area results in more capital cost of network with splitting stream. Network without splitting stream has advantage of economy benefit. However, network without splitting stream has more heat exchangers and is lack of flexibility while network with splitting stream has less heat exchanger and flexibility in operation through adjusting flow rate. Taken consideration of long period investment and flexibility of atmosphere-vacuum distillation equipment, network with splitting stream is selected by most enterprises. The comparisons of heat recovery, area of heat transfer, capital cost, and annual operatering cost of utility between pinch analysis and mathematical programming approach are implemented.The present dissertation takes the flow simulation, optimization and synthesis of heat exchangers network of atmosphere-vacuum distillation as the research. Applied theory of flow simulation, optimization, and synthesis of heat exchangers network to simulate and optimize the distillation process, to integrate heat exchangers network, the optimization flow with maximum annual profit and heat exchanger network with splitting stream and without splitting stream having minimum annual cost are obtained. The comparison of results of pinch analysis and mathematical programming approach are analyzed.
|
PDF Full Text Request