Simulation And Scheduling Optimization Of Atmospheric And Vacuum Distillation Process Based On Asphalt Production

Atmospheric and vacuum distillation is the simplest process for the production of road asphalt with low cost,and around 70%?80%of road asphalt is produced by this process.As the first procedure of crude oil refining process,the comprehensive energy consumption of atmospheric and vacuum distillation takes up a large proportion of the total energy consumption of crude oil processing.The refining and chemical industry is also facing challenges including resource shortage,product quality,environmental protection supervision and economic benefits.How to reduce energy consumption,carbon dioxide emissions and improve economic efficiency of the atmospheric and vacuum process has received much attention from refining companies.Based on the existing research progress of atmospheric and vacuum distillation process,scheduling and heat exchanger network comprehensive optimization,this paper proposed a new multi-objective optimization strategy for atmospheric and vacuum distillation process,scheduling optimization model and comprehensive optimization strategy for heat exchanger network.Process optimization,scheduling optimization and heat exchanger network comprehensive optimization of asphalt production process based on oil mixing refining were conducted,and the detailed contents were listed below.A new production process model was established in this paper based on the production data,rigorous simulation results,crude boiling point(TBP)data and crude oil narrow fraction property data of the specific industrial atmospheric and vacuum distillation unit for asphalt production.Not only the influence of raw materials and product properties,but also that of specific equipment and operating parameters on the total pull-out rate and product distribution were considered in the model,which had better accuracy for the description of the heavy oil refining process.On this basis,the production process model of each refinery of a specific company was obtained by regression.The results showed that the model is more accurate in calculating the total pullout rate and product distribution.This paper investigated the multi-objective optimization of the oil refining process.With the objective of maximizing economic efficiency and minimizing energy consumption of heating furnaces and CO₂ emissions,the optimization of atmospheric-vacuum distillation apparatus was conducted by using a multi-objective genetic optimization algorithm(NSGA-II)coded in Matlab.The results showed that the operating conditions of the initial operating case were not optimal.Under the premise of ensuring that the product specifications could meet the design requirements,economic efficiency of the optimized atmospheric-vacuum distillation apparatus could be increased by 25.71%.To conduct the production scheduling optimization for heavy oil mixing refining process in single refineries and companies including four refineries,optimization model of the nonlinear factory-level production scheduling and the mixed-integer nonlinear company-level production scheduling were established.With the goal of maximizing the economic benefits per tons,the plant-level production scheduling optimization could optimize the mixing refining ratio and provide the optional production scheduling scheme under the constraints such as the price change of crude oil,product price change,market demand change,transportation,inventory and production process conditions.The company-level production scheduling optimization model took factory-level production scheduling model as submodule and was combined with the distribution of raw materials and product distribution model and economic model.According to the market circumstance and the production capacity of each refinery,the company-level production scheduling optimization model optimized the allocation of intracompany resources and products uniformly to make full use of the existing resources,equipment and ultimately improved the overall economic benefits of the company.For single refinery and a whole company including four refineries,the plant-level and company-level production scheduling software were developed using the development tools such as 1STOPT,VC,and Office which achieved practical levels.The plant-level production scheduling software was used to optimize each refinery and the optimized mixing ratio for each refinery was obtained as 0.4(mass fraction of A crude oil,the same below),0.84,0.3,and 1.0,respectively.Accompanying the price change of raw materials,the change of the optimal mixing refining ratios was investigated.With the price increase of crude oil A,the proportion of crude oil A in optimal mixing refining ratio of each plant gradually decreased,but the change trends were different in different plant.Through company-level production scheduling optimization,the appropriate production duty of each refinery and the cross-regional sales condition of asphalt products were determined.The results showed that the production duty of the refinery with poor benefit per tons would be reduced in a constrained market situation.The asphalt market demand of the corresponding sales area could be met through cross-regional deployment,and the total revenue obtained through company-level optimization could be increased by 11.6%compared with factory-level optimization.In this paper,different comprehensive strategies were developed for the comprehensive problem of heat exchanger network of heavy oil mixing device at fixed and variable working conditions.For the heavy oil mixing heat exchanger network under fixed working conditions,a heat exchanger network optimization model based on particle swarm optimization was established.The non-split hierarchical superstructure model was used as the process model of the heat exchanger network.Aiming at the problems of equality and inequality constraints,continuous and non-continuous variables,nonlinear,nonconvex,and discontinuous problems,a new particle swarm optimization algorithm was proposed,which transformed the mixed integer nonlinear programming(MINLP)problem that required double-layer optimization into a single-layer nonlinear programming(NLP)problem.The optimization algorithm could cover the optimal heat exchanger network structure and was simple and easy.In this paper,three typical heat exchanger network examples were optimized.The results showed that the new optimization strategy could effectively reduce the cost of heat exchanger network,and the particle swarm optimization algorithm also had better global convergence characteristics.Considering flexible requirements of the heat exchanger network in variable working conditions,a new optimization strategy of energy integration was carried out by using three extreme operating conditions:maximum heat exchange capacity,cold utility quantity and thermal utility quantity,and the maximum heat transfer condition was selected as the reference condition.Firstly,the optimized heat exchange network and heat exchanger area were calculated.Based on the heat exchange network structure,the other two extreme conditions were calculated step by step.The heat exchanger network of the actual refinery was changed,and the three extreme operating conditions were determined by pinch analysis.The results showed that the order of calculation was from the maximum heat transfer condition to the maximum cold utility condition and then the maximum thermal utility condition,which could minimize the annual cost of each working condition.Compared with other calculation orders,the optimization sequence proposed in this paper was effective.