Demonstrating the Benefit of Multi-Objective Optimization and Clustering for the Design of Waste Heat Recovery Systems

摘要

Escalating energy prices and growting 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 fadlities is aproven method;however, optimizing the waste heat recovery systems (WHRS) can be difficult. Particularly, optimizing WEEKS are difficult when process flows vary with respect to time and/ or there are multiple objectives desired from the WEIRS. 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 clusteringalgorithm to illustrate the cost-benefit of optimising to one objective versus another. In this work, acasestudy, usingmodeled data from an existingcannery,is presented to demonstrate the optimisation of the WHRS forthefadlity. 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℉, for pasteurisation, and then water cools them to 85℉. A WHRS will be optimized for recovering waste heat from the retort pasteurisation spne to heat process hot water. The MOEA will evaluate for economic, energy performance objectives and size restrictions. These goals include: minimising up-front costs, maximising the amount of heat recovered and minimising the floor space required for the system. While optimising the system, the MOEA adjusts for seasonal variability, batch processing variability, and a wide range of potentialfoodproducts. The case study demonstrates that MOEA are useful for illustrating the impact of design objectives and designing WHRS systems.