Synthesis Of Refrigeration System Based On Generalized Disjunctive Programming

Refrigeration system plays an important role in process industries,especially in petrochemical processes.Refrigeration system is a large energy consumer,and its synthesis and optimization can not only improve energy efficiency but also save production costs.Synthesis of refrigeration system has become a research hotspot in the field of process system engineering.The general refrigeration system contains multiple pressure levels,which provides different levels of refrigeration for process streams.In order to ensure reasonable allocation of refrigerants and process streams at different levels and achieve the purpose of reducing system energy consumption,the simultaneous optimization of the cooling system and the heat exchanger network must be considered.This paper mainly focuses on the simultaneous optimization of refrigeration system and heat exchanger network.Mathematical programming is used with a general methodological framework and a mathematical model established.The method mainly contains four steps:(1)Exergy analysis is used to confirm the energy consumption of the refrigeration system;(2)A superstructure of a refrigeration system containing all possible structures is established;(3)A mixed integer nonlinear programming(MINLP)model is established according to the superstructure,and solved with the target of minimizing the total compressor shaft work of the refrigeration system;(4)The feasibility of the results is verified by rigorous simulation.In this paper,an improved superstructure of the refrigeration system in the industrial ethylene plant has been developed,which is more practical and effective.However,the uncertainty of process variables caused by the improved superstructure makes it difficult for modeling.In this paper,a new idea is proposed to solve the modeling problem,with the generalized extraction planning(GDP)method introduced into the modeling process.The Big-M method is used to convert the GDP model into the MINLP expression.Finally,two examples are used to verify the effectiveness and feasibility of the proposed optimization method for the refrigeration system.The results of the two examples show that the proposed method can optimize the refrigeration system and the heat exchange network at the same time.A better optimization result is obtained because of the improved superstructure.And the introduction of the GDP method also effectively solves the modeling problem.