Design And Application Of Nonlinear Predictive Controller For Catalytic Cracking

Petrochemical industry plays an important role in our national economy,which provides all kinds of energy for our country,and has a great influence on the development of agriculture,industry,transportation and national defense science and technology.At the same time,the petrochemical industry is also facing the challenge of reducing energy consumption and improving the utilization ratio of oil resources.At present,the majority of petrochemical enterprises have introduced automation technology,advanced process control technology plays an important role in improving petrochemical production,reducing energy consumption and stabilizing the production process,what is more,it can produce obvious economic benefits.Petroleum catalytic cracking is one of the most important processes in petrochemical process control,which consists of three processes: feedstock catalytic cracking,catalyst regeneration and product separation.The purpose of petroleum catalytic cracking is to produce heavy oil cracking reaction under the condition of high temperature and catalyst.The catalytic cracking unit is composed of three parts: reaction regeneration system,fractionation system and absorption stabilization system.As the core part of catalytic cracking,reaction regeneration system processes crude oil to produce various kinds of light oil products.Therefore,it is of great significance to improve the product yield,product quality and economic benefits of the catalytic cracking process by selecting a reasonable reaction regeneration system control technology.The main contents of this paper are as follows:(1)Reaction process and control technology of reaction regeneration system.At the beginning of this chapter,the application status and problems of reaction regeneration system are analyzed,and the process flow of reaction regeneration system is introduced.Based on the analysis of t increasing economic benefit of the reaction regeneration system,the control technologies of the reaction regeneration system are compared and analyzed.(2)Optimal control of model predictive control and double-layered model predictive control for reaction regeneration system.In this chapter,the traditional model predictive control is introduced,and on the basis of the single layer predictive control,introducing the multi priority steady-state target calculation,double-layered model predictive control is formed by considering the priority of each variable and the constraint conditions.At last,the optimal control of the reaction regeneration system is realized by using the double-layered predictive control method.(3)Optimal control of reaction regeneration system,which is based on particle swarm optimization in double-layered predictive control method.This chapter is based on the double-layered predictive control,in order to avoid the adverse effects caused by the relaxation of operating variables and controlled variables,particle swarm optimization algorithm is used to solve the objective function of economics and the objective function of the dynamic matrix control stage.The control method is used to optimize the reaction regeneration system,the results show that the reaction regeneration system based on PSO-DMPC can track the controlled variables and operating variables.(4)Optimal control of reaction regeneration system based on quantum behaved particle swarm optimization.In this chapter,wo show that the traditional particle swarm optimization algorithm can not guarantee global convergence,quantum behaved particle swarm optimization is proposed,the quantum behaved particle swarm optimization(PSO)algorithm can guarantee the global convergence,and the computing speed is faster.The QPSO-DMPC is used in the optimal control of RRS,and compared with PSO-DMPC and QGA-DMPC.The results show that the control tracking effect of QPSO-DMPC is better than PSO-DMPC and QGA-DMPC.(5)Optimal control of reaction regeneration system,which is based on multi-objective quantum behaved particle swarm optimization.This chapter is based on the basic problems of engineering practice,multiple objectives are considered in the optimal control of RRS,he multi-objective quantum particle swarm optimization is proposed.On the basis of establishing the multi objective function of RRS,multi objective quantum behaved particle swarm optimization algorithm is used to solve it,the results are applied to the double-layered predictive control to achieve the goal of the RRS multiobjective steady-state optimization control.Results shows that the double layer predictive control,which is based on multi objective quantum behaved particle swarm,can track the operating variables and controlled variables of the reaction regeneration system.