Global Optimization Of Heat Exchanger Networks By Differential Evolution Method

Heat Exchanger Networks(HENs) synthesis problem is a typical NP-hard problem, which belongs to mixed integer nonlinear programming range. Further, it is characterized by strong nonlinear, tremendous and irregularity solving domain and lots of local solutions. Consequently, HENs is very hard to optimize because of these features. Conventional optimized method for HENs can be classified two types of Pinch method and Mathematical programming methods. However, in case of pinch method cannot obtain optimal network structure and mathematical method easily trap into local solutions, then, According with the characteristics of the problem, this paper developed Differential Evolution(DE) method to optimized HENs problem, and presented various modified strategies to solve continuous and integer variables optimization as following:Firstly, this paper employed standard DE method to optimize HENs problem through handling HENs integer variables and constraints. Furthermore, under solving model of HENs of DE, it tested the ability of jumping local solution of the method, and analyzed influence between optimized accuracy and populations.Secondly, at the basic of HENs optimization by standard DE, aiming at minimizing total annual cost, it analyzed control parameters influence and then presented two effective search strategies to improve solving efficiency of HENs optimization. Two strategies are sketched as cooperative differential evolution for penalty factor and continuous variables and multi-mutated mechanism with self-adaptive of control parameters.The last of main research in this paper,...