用数学规划法求解换热网络优化问题,当换热网络模型考虑所有分流情况时,会包含大量的连续和离散优化变量,因此对大规模换热网络的优化求解非常困难。相比有分流模型,无分流的网络模型结构简单,变量少,求解难度低。本文以分级超结构模型为基础,建立了无分流换热网络模型,以年综合费用为优化目标,提出用单层改进量子粒子群算法综合无分流换热网络,即在生成无分流换热网络结构变量的同时,生成换热负荷变量。相比有分流同步综合模型,不考虑分流的同步综合模型的复杂度和求解时间显著降低。采用两个典型算例对该方法的有效性进行了验证。
When mathematical programming method is used to solve the optimization of heat exchanger network model in which all of splitted streams are considered, it is quite difficult to solve the optimization problem due to large numbers of continuous and discrete optimization variables. However, the model without splitted stream is simpler and easier to be solved. This paper utilizes stage-wise super-structure model to construct the heat exchanger network without splitted stream. Moreover, aiming at the optimal annual cost, improved quantum particle swarm optimization is utilized to synthesize the heat exchanger network without splitted stream, in which both the variables of heat exchanger network structure and the variables of heat load are generated at the same time. Compared with the model with stream splitted, the heat exchanger network without splitted stream has lower complexity and less solving time. Two typical examples verified the effectiveness of the proposed method.