There are many papers and more than two decades of work on using empirical models to predict electricity loads. The general structure of these types of models are, understandably, quite different. Short- and medium-term load models most commonly relate meteorology and day-of-week parameters to observed loads. The longer-term models also use meteorology/climate as explanatory variables, but typically also include “macro” variables like the long-term evolution of population, number of customers, or economic indicators. TELL provides a framework to blend short- and long-term predictions of electricity demand in a coherent way. Using this unique approach allows TELL to reflect both changes in the shape of the load profile due to variations in weather and climate and the long-term evolution of energy demand due to changes in population, technology, and economics. TELL will have an explicit spatial component relating the predicted loads to where they would occur spatially within a grid operations model.

Within IM3, the TELL model will generate predictions of hourly total electricity load for every county in the Continental United States (CONUS). Predictions from TELL will be scaled to match the annual state-level total electricity loads predicted by the U.S. version of the Global Change Analysis Model (GCAM-USA). The inputs for model training include historical meteorology (e.g., temperature, humidity, etc.) and population at the county level as well as historical hourly time-series of total electricity load for each Balancing Authority (BA) in the CONUS. Inputs to run the model forward in time include future meteorology and population at the county-level as well as future annual total electricity loads at the state-level from GCAM-USA. Ultimately TELL produces hourly time-series of total electricity load for every county in the CONUS that are quantitatively and qualitatively consistent with the future annual state-level electricity loads from GCAM-USA.