In the United States, installation of emission-free sources of power generation, such as wind and solar photovoltaics, has in-creased recently. Unfortunately, these generation technologies present unique challenges to utilities and grid operators because they are variable and not dispatchable. While energy storage can provide capacitance to the system and thereby make renewable power more dispatchable, storage implementation at the munic-ipal scale is poorly understood. This paper examines future ap-plications of energy storage to reduce costs and improve sys-tem reliability for electric utilities at the local level. The city of Austin, Texas was selected as the study area because the city has set aggressive targets of 30-35% of total electricity generation from renewable sources, primarily wind and solar photovoltaics, by 2020. For this analysis, generation assets currently used and those planned for future development by the local utility, Austin Energy, are treated in a unit commitment model using a mixed integer programming (MIP) approach. The model has been de-veloped such that it can be provided any objective function and generation portfolio, and the results can be used by whatever stakeholder has generated those particular inputs. To best sim-ulate operational dispatch conditions, the model includes ramp rate constraints, generator turn-on penalties, and minimum oper-ation levels. Energy storage is included by allowing the model to assign an unconstrained asset throughout the study period, 24hours, to whatever values will minimize the objective function. For this initial analysis, storage system efficiency, capital and marginal costs were not included, though they may significantly affect total allocated storage. Modeling results indicate that stor-age availability yields a reduction of as much as $600,000/day in marginal costs for the study area, based primarily on improved utilization of more efficient generating units. This result does not consider savings associated with NO_x reductions. Such re-ductions were studied with a second objective function. While NO_x reductions of approximately 9-23% were observed, these emission reductions were accompanied by significant increases in operating costs. Energy storage requirements and potential cost savings under the scenarios examined might be beneficial to researchers interested in grid-scale storage. These results can also be used to determine appropriate cost targets for storage researchers and manufacturers.
展开▼
