In this paper, an efficient as well as reliable approach to deal with heat exchanger networks (HENs) synthesis problems, which is inherently known as a mixed-integer non-linearprogramming model, is presented. The str...
详细信息
In this paper, an efficient as well as reliable approach to deal with heat exchanger networks (HENs) synthesis problems, which is inherently known as a mixed-integer non-linearprogramming model, is presented. The structural variables as the discrete variables are optimized by a genetic algorithm (GA), whereas continuous variables are handled by a modifiedquasi-linearprogramming (MQLP) model. Each HEN is considered as a chromosome consisting of a sequence of genes. Each gene also contains the address of the exchanger(s) in the network. The HENs generated by the GA are sent to the MQLP to calculate their overall objective function (OOF) (i.e. minimum total annual cost (TAC)). The MQLP model includes two inner and outer surfaces. On the outer surface, the local optimal values of the continuous variables are found according to the maximum energy recovery of HEN, while on the inner surface, the globally optimal values of them are found to reach the minimum TAC of HEN. Due to the relatively linear behavior of the proposed method, a comparison of results with references showed that this method can reduce TAC of HENs compared to the studied references by about (0.51% to 2.37%). (c) 2021 Elsevier Ltd. All rights reserved.
暂无评论