Between increasing demand driven by AI and data centers, supply chain and tariff issues that confound necessary infrastructure developments, and progressively more volatile weather patterns and temperature extremes, it’s harder than ever to match supply and demand. Last May, the U.S. Energy Information Administration reported that electric demand is rising after a decade of stagnation, with further analysis suggesting that demand could grow by 25% by 2030. To keep up with rapidly rising demand, utilities leverage distributed energy resources (DERs) for demand flexibility initiatives like virtual power plants, demand response, and EV charging. Their secret on when and how to best deploy those strategies: energy forecasting.
Why Energy Forecasting Matters
Energy forecasting is the practice of using historical and/or real-time data to inform future energy needs with the intent to match supply and demand, while remaining competitive in energy markets. Ideally, energy forecasting is useful in energy arbitrage practices that focus on buying energy at lower costs to deploy during peak-periods of demand, typically realized through DERs like battery energy storage systems (BESS).
As weather patterns continue to evolve and demand rises, forecasting provides an opportunity for program managers and grid operators alike to plan accordingly for their future needs, whether that’s to purchase the necessary energy needed to meet demand in advance or to run demand flexibility initiatives meant to curtail or offset usage during peak periods of demand through aggregate conservation efforts. Altogether, forecasting efforts are effective in helping mitigate wholesale energy prices, operational costs, and enhance grid resiliency.
Types of Forecasting
As with any statistical model, energy forecasting can manifest in many ways. As noted, energy forecasting is useful as a means of preparing for uncertain electric demand. While the volume of potential of daily energy need may vary, forecasting is an excellent tool to optimize energy purchasing strategies to mitigate costs, while preparing for grid events that may challenge the match between supply and demand. In general, forecasting might look at short, medium, or long-term forecasting models, may break down methodology by statistical models, AI/machine learning, physical or hybrid models, or may vary based on what is being forecasted (load, renewable energy, costs, etc.). For this article, we’re comparing system load forecasting and DER event forecasting.
System Load Forecasting
As the name suggests, load forecasting is the analytical process of assessing available load for future events, outcomes, or trends to determine the aggregate system load for a utility or region. Based on historical, real-time, and probabilistic data, system load forecasting is designed to simplify the complexity of challenges that come from traditional modeling capabilities. Event load forecasting considers both erratic customer behaviors as well as volatile and evolving weather patterns to predict future demand. This is useful in grid planning, as well as in informing energy market purchases.
DER Event Energy Forecasting
Unlike event load forecasting, DER event forecasting specifically looks at the collective potential outcome of DERs for use in demand flexibility programs by predicting the baseline DER load and potential grid event performance for DERs. As such, DER event forecasting is an effective strategy to determine the potential opportunities that DERs can provide during grid events.
DER event forecasting cuts through the challenges of behind-the-meter DERs like solar, battery energy storage systems (BESS), electric vehicles, EVSE chargers, and smart home devices like thermostats or water heaters are subject to customer activity during grid events like customer behaviors, to provide an accurate picture of potential, while proactively informing internal decision-making across departments.
The Value of a Comprehensive Forecasting Strategy
Right now, utility customers have two things on their minds as it regards U.S. electric demand: will power stay on, and will it be affordable? Energy forecasting helps mitigate these challenges by providing a path for how to most effectively manage electric demand and market costs, while helping utilities make informed decisions on necessary, but incredibly costly, infrastructure upgrades. Individually, system load forecasting is an effective tool to determine potential needs, while DER event forecasting helps inform how and when to run demand events; both support energy market purchasing teams.
By leveraging multiple types of energy forecasting, utilities can create a comprehensive prediction of likely energy demand needs. When combined with a Grid-Edge distributed energy resource management system (DERMS), which aggregates otherwise disparate behind-the-meter DER assets for use in demand flexibility programs, AI, and model predictive control, forecasting tools can be used to foster accurate outcomes from customer demand flexibility programs.
Functionality like Topline Demand Control (TDC) enables the next generation of virtual power plants by optimizing DERs at a granular level to ensure a guaranteed outcome. Put differently, by leveraging forecasting software with DERs, utilities can rest easy knowing that they will get a desired outcome from BTM DERs without any guesswork.
The Difference Between Load & DER Event Forecasting: Conclusion
Electric demand is exponentially rising, causing rapid demand growth, while drawing consumer scrutiny. Because of this market uncertainty, the electric demand and supply market for forecasting is rapidly growing: enterprising utilities are already working hard to better predict an increasingly uncertain energy future, and energy forecasting techniques like system load forecasting and DER event forecasting are critical in creating a broader, more nuanced image of potential energy needs.