Nonlinear Optimization in Equipment Size Selectionfor Distributed Generation Applications

摘要

Evaluation of potential distributed generation combined heat and power (CHP) applications requires an assessmentrnof the operations and economics of a particular system in meeting the electrical and thermal demands of a specificrnend-use facility. Software is currently available that will simulate a given CHP system in a specific application.rnOperation simulation is performed on an hourly basis which reflects diurnal loads. CHP equipment capacity is sizedrnin these programs heuristically on the basis of either annual peak electrical or thermal load. Each simulation runrnproduces a single-point scenario for a CHP system. The weakness of this approach is that there are multiplerntechnical and cost impacts related to the size of the CHP system that influence the overall project viability.rnDetermination of the most optimal capacity for the CHP system requires an iterative process, as the number ofrninfluences and their timing produce a non-linear response to equipment size.rnThis paper addresses research being conducted on selecting nonlinear optimization algorithms for optimizing therncapacity of combined heat and power systems. Optimization is based on maximizing cost savings from the use of arnCHP system relative to meeting the electrical and thermal demands from traditional utility grid and gas-fired boilerrnsources. After evaluating the behavior of the objective function under differing input scenarios, line search methodsrnthat do not rely on objective function derivatives appear to be most effective in finding a global optimum capacityrnsize. The findings from this research have been incorporated into a PC-based computer program to perform suchrnoptimizations.