How AI is reshaping DER potential studies
The energy landscape is undergoing a significant transformation, shifting from traditional centralized power systems to a more dynamic and distributed approach. Distributed Energy Resources (DERs) are at the forefront of this evolution, providing utilities with the opportunity to enhance grid reliability and reduce costs.
Historically, utility programs were designed to address broad, macro-level energy challenges focused on state or province-wide needs rather than localized issues. Some customer programs such as DERs can address localized needs, reducing the need for significant infrastructure investments and preventing upward pressure on retail rates.
With advancements in AI, energy analytics, and grid monitoring, utilities can now tailor all customer programs to specific geographic areas, treating customer energy resources as local assets. Implementing localized solutions helps utilities to:
- Enhance grid reliability by strategically deploying DERs in areas with known grid constraints, ensuring efficient energy distribution.
- Reduce costs by prioritizing investments where they are most needed and leveraging customer participation in energy programs to offset infrastructure expenses.
- Improve efficiency and accessibility by ensuring that energy programs are available to all customers.
Localized programs are becoming increasingly important to utility planners, who rely on forward-looking analyses like potential studies to reflect evolving program designs and accurately project savings. Addressing this demand requires advanced analytics and data science to deliver highly granular forecasting. As the need for analytical precision grows, the strategic value of potential studies increases, making experienced providers essential for generating actionable, data-driven insights.
Evolution of potential studies
ICF has been conducting demand-side management potential studies for utilities and government agencies across the world for more than 20 years. These analyses provide an estimate of the amount of energy savings or load reductions that can be achieved through various strategies.
Initially, utilities were primarily interested in energy efficiency and limited demand response (DR) programs. As programs and technologies evolved, utility interest in more advanced DR programs, electrification technologies, and assets such as solar and storage increased. This added the first layer of complexity for potential studies.
As shown in Figure 1, our team developed a model to meet these evolving needs of utilities. This model evaluates the potential of various behind-the-meter resources using a unified framework. It enhances the consideration of interactive effects among different resources, ultimately resulting in a more consistent and accurate forecast of DER potential.
Figure 1: Unified DER planning potential model
Built upon the principles outlined in the National Action Plan for Energy Efficiency guide, the model applies a bottom‑up approach to estimate different levels of potential based on stock turnover. It’s been extensively tested and validated, and it has undergone both third-party stakeholder review and scrutiny by Public Utility Commission staff in several jurisdictions.
As electricity demand continues to rise—projected to grow 25% by 2030 and 78% by 2050 compared to 2023—utilities are shifting focus to more localized challenges, such as transformer overloads. These issues require targeted, customer-level analysis to manage load growth effectively. This adds a new layer of complexity to potential studies, which must now support substation-level planning.
At the same time, rapid growth in DER adoption presents its own set of analytical challenges. Accurately forecasting and optimizing DER deployment demands sophisticated modeling and highly granular analysis capable of capturing customer-specific impacts on the grid. Traditional potential study frameworks often lack both the resolution and localization required to assess DER impacts, limiting their ability to support effective grid solutions.
Figure 2: Power of granular forecasting
AI technology—which uses data to create a virtual representation (or digital twin) of a product, asset, or system—presents a robust solution to address these challenges by simulating real-world grid conditions and customer behaviors, leading to better decision-making and enhanced grid resilience. 
Sightline, ICF’s end-to-end utility customer program platform, uses AI to generate calibrated digital twin models and highly granular forecasts (e.g., detailed building-level simulations and hourly load shapes) to create a more accurate view of energy performance throughout the year. This approach enables utilities to evaluate savings potential for both energy and demand across all DSM technologies with greater precision.
Sightline also calibrates the digital twin by ingesting large datasets for all customers in a utility company’s territory, such as hourly AMI (Advanced Metering Infrastructure) data when available. We utilize that granular AMI data for targeted modeling, which allows utilities to tailor customer programs to specific geographic areas and operational needs. This level of precision becomes especially valuable when leveraging digital twins to address localized grid constraints.
Using digital twins to address localized grid constraints
Digital twins represent the most accurate simulation of a real building, enabling precise evaluation of the effects of various energy technologies. Our methodology for creating digital twins incorporates principles of heat transfer and thermodynamics to develop a deterministic model that reliably forecasts the impact of these technologies. This process is computationally intensive and has only become viable with recent advancements in AI and cloud computing.
Digital twin technology leverages both physics-based models and machine learning approaches to optimize forecasting accuracy. This foundation enables the following key steps for modeling DER potential, as shown in Figure 3.
Figure 3: Key steps in using digital twins to model DER potential
Benefits of integrating digital twins with potential studies
With a digital copy of a building via digital twin technology, utilities can simulate the impact of DERs to understand the real-life effects. This allows them to:
- Model DER adoption and its impacts at the distribution level, enabling precise grid planning. 
- Identify existing and emerging grid constraints, allowing for proactive infrastructure investment. 
- Forecast premise-level load variations, ensuring that supply and demand are balanced efficiently.
- Simulate different DER adoption scenarios, providing insights into the most effective incentive structures and program designs.
- Assess flexible load management strategies, optimizing demand response programs and time-of-use pricing to enhance grid efficiency. 
This level of modeling precision is now being operationalized in utility planning and is actively shaping large-scale efforts across the energy sector, as shown in the following case study.
Case study: Using digital twins to forecast DER potential at a substation level
We worked with an independent system operator in North America to simulate its baseline usage for calibration against historic and current data, and then to apply any customization needed to represent its local building stock.
This first-of-its-kind potential study for a system operator in North America used digital twins to forecast DER potential at a substation level for 25 years. The study included potential analysis for energy efficiency, demand response, and behind-the-meter PV and storge. The analysis was done at the building level and was aggregated up to the substation level to support their regional planning for non-wires alternatives.
The path forward
The future of DER adoption and grid modernization depends on accurate, granular, scalable forecasting methodologies. Digital twin technology, like that in our Sightline platform, offers utilities a powerful, data-driven solution to optimize DER deployment and enhance grid resilience. By leveraging granular forecasting, advanced modeling, and localized insights, the energy sector can unlock the full potential of DERs.