Simultaneous Synthesis And Retrofit Of Heat Exchanger Network Considered Pressure Drop

The simultaneous synthesis and optimal retrofit of Heat Exchanger Network have been the important research areas of process system energy integration, it acts as an important role to raise the energy utilization ratio of process industry. The rationality of the design, operability, and whether the optimization has great influence on energy consumption of the process system. During the past few decades, heat exchange network synthesis and retrofit has made great progress, and the research results have been widely applied in the actual process of industrial system and create the considerable economic benefits. However, for a lot of factors which always affect the heat exchange network energy consumption and economic benefits, such as dirt, pressure drop and different material of heat exchanger, the mathematical programming method which based on the stage-wise superstructure still don’t has a comprehensive consideration on them. Therefore, the research content of this paper is based on the non-isothermal mixed stage-wise Heat Exchanger Network superstructure, and a synthesis model of Heat Exchanger Network with pressure drop consideration is established. This model will be applied to the simultaneous synthesis Heat Exchanger Network which considers the pressure drop, and after simple improvement it can be applied to the retrofit of a existing Heat Exchanger Network. Finally, the genetic/simulated annealing algorithm will be used to solve the mathematical model present by this paper.This paper is concentrate on the simultaneous synthesis and optimal retrofit of Heat Exchanger Network which considers pressure drop, based on the non-isothermal mixed stage-wise Heat Exchanger Network superstructure, the work in this paper is as follow:(1) The synthesis model of Heat Exchanger Network with pressure drop consideration is established, and the minimum total annual cost as the objective function. Based on the non-isothermal mixed stage-wise Heat Exchanger Network superstructure presented by Bjork and Westerlund, the constraint conditions about the pressure drop and flow velocity have been added. Namely, the velocities of cold/hot processes are act as new optimization parameters in the model. The non-isothermal mixing is been involved, and this way is more close to actual situation. A MINLP mathematical model about simultaneous synthesis Heat Exchanger Network considers pressure drop is been present.(2) Used the GA/SA algorithm to solve the MINLP model. The GA/SA algorithm is combined with genetic algorithm and simulated annealing algorithm, and it will be used to solve the mathematical model present by this paper, and the produce of the initial solution, operator design, the constraint conditions have been studied. Through two examples to verify this model, and compared to literatures, the annual total cost can decrease34.9%and3.2%respectively, the results proved that the model considering pressure drop and the GA/SA algorithm can get better solution. The pressure drop of the network will be considered simultaneously with the other three factors:number of heat exchangers, heat transfer areas and the utility consumption, after the pressure drop values are optimized, the pump costs compared with the literatures decrease69.8%and19.1%respectively, so the model considering pressure drop will have a large savings in the power cost.(3) The retrofit MINLP model of Heat Exchanger Network considering pressure drop has been established, it is derived from the synthesis model of Heat Exchanger Network with pressure drop consideration. But it adds some constraint conditions about the Heat Exchanger Network retrofit into the synthesis model, so a retrofit model is presented. Applied this model to an example, the results of the calculation about utility saving is equivalent to78.0%of the literature, but the investment about heat transfer area only accounts for35.9%of the literature. It’s indicate that the model present by this paper has reached the goal, less investment costs but more utilities to save. The feasibility and validity of this model is verified.