In a new report, experts from The Brattle Group examine the rapidly changing landscape of load drivers and how they complicate load forecasting for utilities and system operators. The authors find that improved forecasting – the projection of future peak demand and energy consumption – with the inclusion of these new drivers will be an essential component in the transition to a decarbonized electricity system that is both reliable and affordable.

The report, Electricity Demand Growth and Forecasting in a Time of Change, provides an overview of several new and emerging demand drivers that will affect the growth and patterns of load in the coming decades and the ways in which utilities are currently including them in their forecasts. The new drivers include data centers, the expansion of manufacturing and industrial processes, the electrification of transportation and buildings, and cryptocurrency mining. The potential impact of demand-side resources – including distributed generation (DG), energy efficiency (EE), and demand response (DR) – that can partially offset these load drivers is also considered.

“Currently, there is a wide spectrum among utilities in how they account for these new drivers,” said Brattle Principal T. Bruce Tsuchida. “The future net load growth spurred by the new drivers is vast, and our analyses suggest that – given this growth, along with the change in load characteristics and other associated uncertainties – the industry will require a revamped approach to load forecasting moving forward.”

Notable findings from the report include:

  • The combined speed and magnitude of the new growth drivers point towards a sustained period of high electric demand growth for many parts of the US, though it is likely that growth will vary substantially across regions and even within a utility’s footprint. This highlights the importance of longer-term forecasting specific to each area’s circumstances.
  • The potential to moderate the growth of both electricity consumption and peak load through DG, EE, and DR programs is also material; however, the degree of moderation will likely vary substantially based on the programs implemented in a given area. With the expected net load growth, advanced and granular forecasting that fully includes these resources will be necessary.
  • Though public documents show that many utilities have made significant progress in incorporating these changing drivers into their forecasts, very few forecasts appear to reflect the full set of new drivers.

The report also includes a discussion of the costs of over- and under-forecasting load, both individual drivers and as a whole. The authors find that, in today’s world – where many of the new drivers are policy-driven (by, for example, decarbonization targets or industry onshoring policies), the costs and long-lasting impact of under-forecasting are much larger than those of over-forecasting.

Electricity Demand Growth and Forecasting in a Time of Change was coauthored by Principal T. Bruce Tsuchida, Managing Energy Associates Dr. Long Lam and Akhilesh Ramakrishnan, Senior Advisor Peter Fox-Penner, Energy Specialists Sylvia Tang and Adam Bigelow, and Senior Energy Analyst Ethan Snyder. The full report can be found here.