Design optimization of industrial installations with help of mathematical programs offers high potential for energy savings and cost reduction, which steadily gains importance. Heat exchange network synthesis (HENS) is one of the commonly used and promising methods. This paper deals with simultaneous integration of heat pumps and two different storage types into mixed integer linear programming (MILP) HENS by extending a multi-period superstructure formulation proposed by Beck and Hofmann (2018c). Further measures for tightening the solution space for the coefficient of performance (COP) are introduced. This improves the accuracy of the heat pump (HP) approximation, reduces computational effort and prevents solutions with uneconomical low COP. The obtained approach allows to design cost efficient heat recovery systems with storages (ST) and HP for the improvement of the energy recovery capabilities. A constructed test case was used to analyze the performance of the method and to show its possible potential.

在数学程序的帮助下，工业装置的设计优化为节能和降低成本提供了巨大的潜力，而这一潜力日益重要。热交换网络合成（HENS）是常用且有希望的方法之一。本文通过扩展Beck和Hofmann（2018c）提出的多周期上部结构公式，将热泵和两种不同的存储类型同时集成到混合整数线性规划（MILP）HENS中。引入了进一步的措施来缩小性能系数（COP）的解决方案空间。这提高了热泵（HP）逼近的精度，减少了计算工作量，并防止了具有不经济的低COP的解决方案。所获得的方法允许设计具有存储器（ST）和HP的经济高效的热回收系统，以提高能量回收能力。使用构造好的测试用例来分析该方法的性能并显示其可能的潜力。