PJM released the 2016 Load Forecast Report on December 30. This report is based on a significant revision to the forecast model since last year’s report.
The primary revisions reflect new trends in energy efficiency and weather measurements in the peak demand forecast. These revisions were reviewed with stakeholders and endorsed by the Markets & Reliability Committee.
The report includes long-term forecasts of peak loads, net energy, load management and energy efficiency for each PJM zone, region, locational deliverability area and the total RTO.
- Adding variables to account for trends in equipment and appliances saturation and efficiency
- Revised weather variables; examining 20 years of weather patterns instead of 40 years
- Updated economic area/weather station assignments to zones
- Revised weather normalization procedure
“We restructured how we treat weather to allow for more variable load responses across a wider range of conditions,” said Tom Falin, manager – Resource Adequacy Planning. “We added three variables – cooling, heating and other – to account for trends in equipment and appliance saturation and efficiency.”
Falin added that distributed solar generation is now reflected in the historical load data used to estimate the models, with a separately derived solar forecast used to adjust load forecasts.
In all load models, PJM used historical data from January 1998 through August 2015. PJM staff used weather data from 1994 through 2014 in simulating its models, generating 273 scenarios.
Other information components:
- PJM used Moody’s Analytics’ October 2015 release for the economic forecast.
- Equipment indexes reflect the 2015 update of Itron’s end-use data, which is consistent with the Energy Information Administration’s 2015 Annual Energy Outlook.
Falin said PJM will publish a white paper on the forecast model in 2016 providing background and greater detail on the new model.