Improvement On Operation Optimization Of Petrochemical Unit Aided With Process Simulation Technology

Due to the changes in the nature of raw materials,product requirements and other factors,the actual operation of petrochemical production unit will deviate from the design condition,causing extensive energy consumption and worse product distribution.How to adapt to the change of working conditions and make the unit run in an ideal state without changing the process and equipmentby adjusting the adjustable parameters has always been a problem of high concern in the petrochemical industry.Therefore,the optimization of multivariable operation in petrochemical process based on simulated data is implemented in this paper.Main research:1.Sensitivity analysis,range analysis,and variance analysis are used to select excellent operating variables in order to strengthen the impact on the objective function and improve the convergence of the simulation because of the numerous operation variables of petrochemical process,their respective different mechanism of action on the target function and effects of PRO/II and ASPEN simulation rates.2.The optimization constraint is effectively strengthened and enriched,including additional intra-device constraints such as thermal binding and thermal discharging constraints and strengthening constraints of product quality,machine pump load and simulation convergence property.The irrational simulation calculation is largely circumvented,the optimization area is narrowed,the calculation efficiency is improved,and the optimal solution is closer to the reality.3.The genetic algorithm(GA)has a strong global optimization ability and is used as the optimization calculation method to optimize the multi-variable operation based on data.But in the process of implementation,there are prevalent problems of large data shock,slow simulation,or even search stagnation.In this paper,the small area search method with the current optimal value as the center is proposed,and then it is transferred to the next generation after obtaining the better value of the small region,so as to gradually approach the global optimal value.Other improvements have been made,including: 1)The elite strategy is adopted to improve the fitness recession of the optimization process;2)In the following generation,the same individuals as the previous generation are directly assigned to the same value to avoid repeated calculation;3)The selection operation of particle swarm optimization(PSO)is used to replace the selection operation of genetic algorithm in order to improve its local optimization ability.The improvement in four aspects can increase GA calculation speedby at least 2 times.4.The PRO/II file is invoked under the Excel interface through programming Excel VBA.This method not only improves the efficiency of simulation,but also provides friendly operation interface for users.At the same time,in VBA programming,the genetic algorithm for the implementation of the above improvement is embedded,and automatic data communication between PRO/II and Excel is applied to realize the multi-variable automatic optimization of the complex petrochemical process.Through the above improvement,the multi-variable operation optimization method of complex chemical process based on simulation data is faster,better and more reliable.The application of 800,000 tons per year wax oil catalytic cracking unit 's main fractionating tower and its absorption stability unit validates the feasibility of the method.when the extracted volume in the first middle section,the flow of the lean diesel and supplementary absorbent are selected as optimization variables,it gets the best combination of variables after only 13 generation of genetic algorithm optimization.Compared with the existing operating conditions,this method increases economic benefits of 14.61 million yuan per year.