Dynamic Models for Wind Energy Diffusion and System Integration: Aspects of Demand Response to Mitigate Forecasting Errors

摘要

Demand Response is becoming a substantial contributor to reliability of the electric grid in the United States. Today, Demand Response has the potential to alleviate GW worth of capacity constraints on the grid nationwide. Concurrently, growth of wind electricity generation capacity has caused concern due to the variable nature of the resource. Coupling the benefits of Demand Response to wind provides opportunities to create a greener grid that is just as stable as its fossil-fuel based counterpart. This paper shows how Demand Response can theoretically and empirically work with wind forecasting to enhance grid reliability and improve profits to wind facilities. This paper demonstrates how Demand Response can be used to 1) mitigate the reliability issues associated with forecasting error for wind generation and 2) offer a potential mechanism to increase overall profits to wind-generation facilities. In the first case, extreme wind events are known to cause "ramp" issues for the grid where the wind resources available change dramatically in a short time period. Large-scale peaking plants may have trouble accommodating the changes on the time-scale of interest (within an hour time frame). Demand Response can be actuated within an hour time frame thus providing badly needed stability to the grid under extreme weather conditions. Secondly, by serving as a "call-option" to the wind industry, Demand Response may actually allow for increased profits due to elimination of the downside risk of underestimation in wind forecasts. All of this analysis, of course, assumes a regulatory structure in the utility markets that would allow these mechanisms to be used. In this paper, a simple forecasting process, using quantile regression, is used for an extremely small (< 10 MW) wind farm for an independent municipal utility. It is then shown how Demand Response could theoretically and empirically improve the overall reliability and profitability of the system.