不断上涨的能源价格和消费者对可持续产品的日益关注促使制造部门减少能源消耗。尤其是，食品生产行业需要大量热能来烹饪食品，其中9%到12%通常被浪费掉。回收和再利用设施内的废热是一种行之有效的方法；然而，优化余热回收系统（WHR）可能很困难。尤其是，当工艺流程随时间变化和/或WHR需要多个目标时，优化WHR是困难的。复杂系统可以通过使用多目标进化算法（MOEA）进行优化。然后，可以使用聚类算法分析多目标设计空间的权衡，以说明优化一个目标与另一个目标的成本效益。 在这项工作中，使用现有罐头厂的建模数据进行了一个案例研究，以证明该设施WHR的优化。罐头厂按季节运作。在操作过程中，生蔬菜和肉被清洗、煮熟、调味和罐装。一个特别耗能的设备是蒸馏器。干馏蒸汽将罐加热至160°F进行巴氏杀菌，然后用水将罐冷却至85°F。WHR将被优化用于回收干馏巴氏杀菌区的废热，以加热工艺热水。经济部将评估经济、能源绩效目标和规模限制。这些目标包括:最大限度地降低前期成本，最大限度地回收热量，并最大限度地减少系统所需的占地面积。在优化系统的同时，经济部根据季节变化、批量加工变化和广泛的潜在食品进行调整。 案例研究表明，MOEA有助于说明设计目标的影响和WHR系统的设计。引用:2019年冬季会议，佐治亚州亚特兰大，会议论文

Escalating energy prices and growing consumer concern for sustainable products incentivizes the reduction of energy consumption within the manufacturing sectors. Particularly, the food production industry requires large amounts of heat energy to cook food, of which 9 to 12% is typically wasted. Recovering and reusing waste heat within facilities is a proven method; however, optimizing the waste heat recovery systems (WHRS) can be difficult. Particularly, optimizing WHRS are difficult when process flows vary with respect to time and/or there are multiple objectives desired from the WHRS. Complex systems may be optimized by using a multi-objective evolutionary algorithm (MOEA). The multi-objective design space tradeoffs may then be analyzed using a clustering algorithm to illustrate the cost-benefit of optimizing to one objective versus another. In this work, a case study, using modeled data from an existing cannery, is presented to demonstrate the optimization of the WHRS for the facility. The cannery operates seasonally. During operation raw vegetables and meat are cleaned, cooked, seasoned and canned. A particularly energy- intensive piece of equipment is the retort. The retort steam heats cans to 160°F, for pasteurization, and then water cools them to 85°F. A WHRS will be optimized for recovering waste heat from the retort pasteurization zone to heat process hot water. The MOEA will evaluate for economic, energy performance objectives and size restrictions. These goals include: minimizing up-front costs, maximizing the amount of heat recovered and minimizing the floor space required for the system. While optimizing the system, the MOEA adjusts for seasonal variability, batch processing variability, and a wide range of potential food products. The case study demonstrates that MOEA are useful for illustrating the impact of design objectives and designing WHRS systems.