特殊换热网络算例优化与分析

Optimization and analysis of specific heat exchanger network cases

换热网络综合属于混合整数非线性规划问题,具有严重的非凸、非线性特性。为探讨某些换热网络算例可能存在内部公用工程且年综合费用极低的特殊换热网络结构,本文采用了允许公用工程内置的无分流分级超结构模型。基于布谷鸟搜索算法,通过设置最小换热量阀值及以较小概率接受较差结构来指导换热网络结构进化,建立适用于求解换热网络综合问题的优化算法,并对特殊算例的优化结果进行分析。本文列举3个特殊算例,具有热流体的初始温度高于热公用工程的入口温度或热公用工程的进出口温差较大的特征,改进的布谷鸟搜索算法能够搜索到带有内部公用工程的换热网络结构,且年综合费用得到明显下降。结果表明:对于特殊算例,公用工程内置能够优化换热网络的热量分布,节约换热器面积,有效降低换热网络的年综合费用。

Heat exchanger network synthesis belongs to the mixed integer nonlinear programming problem with severe nonconvex and nonlinear characteristics.In order to study some heat exchanger network cases probably existing special heat exchanger network structures with inner utilities but with lower total annual cost,a non-splits stage-wise superstructure with inner utilities was utilized.Based on the cuckoo search algorithm,an optimization algorithm was established by setting the minimum heat threshold and accepting bad network structures with a small probability to guide the structure evolution process and applied to solve heat exchanger network synthesis problems,and optimization results of the specific cases were analyzed.Three specific cases,with the characteristics of the initial temperature of the hot stream is higher than the inlet temperature of the hot utility or the large temperature difference between the inlet and outlet of the hot utility,were solved with the modified cuckoo search algorithm which found the heat exchanger networks with the inner utility and those total annual costs have a steep descent.For specific cases,the results showed that the inner utility can serve as an important role of optimizing the heat load distribution of the heat exchanger network,which save the heat transfer area and effectively reduce the total annual cost of the heat exchanger network.