VTechWorks staff will be away for the Thanksgiving holiday beginning at noon on Wednesday, November 27, through Friday, November 29. We will resume normal operations on Monday, December 2. Thank you for your patience.

Browsing by Author "University of Maryland, College Park. Department of Atmospheric and Oceanic Science"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Modeling the variability of renewable generation and electrical demand in RTOs and cities using reanalyzed winds, insolation and temperature
    Kirk-Davidoff, Daniel; Jascourt, Stephen; Cassidy, Chris (Virginia Tech, 2015-06)
    Wind and solar generation forecasts are used by electrical grid operators and utilities to predict the amount of electricity that will be required from non-renewable generation facilities. That is, they are typically regarded as contributing to a prediction of effective electrical demand. Electrical demand itself varies strongly and non-linearly with temperature. Thus skillful prediction of effective electrical demand requires good predictions of both demand-weighted temperature and renewable generation. Since electrical prices depend not only on demand, but on expectations of demand, the goal of avoiding spiky electrical prices and reduction of the use of inefficient peaking power units is best served when the joint variability of demand and renewable generation is well understood. In this presentation we examine the joint variability of modeled electrical demand and wind and solar generation in all the US and Canadian regional transmission organizations (RTOs). We use reanalyzed wind, temperature and insolation from the Climate Forecast System Reanalysis to simulate the demand and renewable generation that would have resulted over the past 33 years if the present renewable infrastructure and temperature-sensitive demand had been present over that period, and then extend the present infrastructure to simulate variability at 35% renewable contribution to electrical demand. Demand sensitivity is based on regression of temperature and electrical demand over the past year (2013) in each RTO. We show that the correlation of demand and renewable generation is a strong function of geography, season, and averaging period (hourly, daily or monthly), identifying those regions where joint variability tends to reduce effective demand variability and those where joint variability tends to increase effective demand variability. While wind generation typically maximizes at night, the intensity of the diurnal cycle varies strongly over the course of the year, while temperature dependent demand variations and solar generation variations are largest in summer, since temperature is more diurnal at that time than in the winter, and the sensitivity of demand to temperature is larger in summer. We explore the ways in which choices of how and where to expand renewable generation affect the co-variability of renewable generation and demand, and the probability of long periods of low generation in each RTO. We anticipate that this simulation capability will be helpful in planning the needs of the future electrical grid for generating capacity, storage and demand management, as well as for inter-regional power transfers. Wind, solar and demand co-variability differ strongly as a function of geography. The appropriate balance of wind and solar generation in an RTO should be determined not only on the basis of the mean resource, but also with the goal of robust operation, by minimizing the probability of long periods of simultaneous low renewable generation and high electrical demand.