Quarterly Meeting Recap

Q4 OpenEAC Alliance Meeting

Thanks to everyone who attended the quarterly meeting of the OpenEAC Alliance.

The slides can be found here: https://docs.google.com/presentation/d/1JVaFM8MB0ZnlPIVJwXC_pnd8qou6CbAYbSYQ2jI_QlQ/edit?usp=sharing

A full recording can be found here: https://app.fireflies.ai/view/OpenEAC-Alliance-Quarterly-Call::01KA1W0E94J2KJR78JH3BYQ010

The Scene from Real Genius can be found here

Next Meeting: February 19th, 2026

Summary:

• Demand Response Methodology Finalized: Standardized method for short-term demand response savings using a 28-day baseline with hourly data.

• Flexibility in Methodology: M&V plans can adjust baselines to include or exclude prior events, improving accuracy and adaptability.

• Open Source Collaboration: Methodologies available at methods.openac.org for public feedback, fostering transparency and iterative improvements.

• Market Confidence: New methodologies aim to accurately measure demand-side energy reductions, boosting confidence in distributed energy resource value.

• Multi-Attribute Measurement: Savings expressed in energy, carbon, dollars, and capacity to meet diverse valuation needs and support emissions reduction.

• Continuous Development Planned: Future expansions include batteries, hybrid generation, and enhanced carbon accounting efforts, with community engagement ongoing.

AI Notes

Demand Response Methodology

The meeting finalized a standardized, transparent methodology for measuring short-term, event-based demand response savings using a 28-day baseline and hourly data.

• The demand response approach uses hourly whole-building electricity data and localized temperature data to model counterfactual consumption during rare demand response events, typically lasting a few hours, to calculate savings (16:15)

  • The model is based on OpenDSM’s eemeter regression, adapted to shorter baselines suited for event windows rather than year-long data typical for energy efficiency.

  • A 28-day baseline was chosen to capture recent weather and consumption patterns conservatively, with room for adjustment in special cases like shoulder seasons.

  • Events should be rare; frequent load shifts are excluded and handled by a separate load shifting methodology.

  • Bounce-back effects after the event are not included as this method focuses solely on the event period’s grid impact.

• Methodology flexibility allows M&V plans to decide whether to include prior events in the baseline period, accommodating edge cases and varying regulatory needs (22:29)

  • Including prior events in the baseline accepts that savings from sequential events influence each other.

  • Excluding them might better isolate specific event impacts depending on the use case.

  • This flexibility supports broader applicability without sacrificing methodological rigor.

• Open source publication and collaboration were emphasized by McGee Young, with all methodologies and updates available at methods.openac.org and GitHub for version control and transparency (12:22)

  • The Open EAC Alliance encourages public comment and iterative improvement through this peer review approach.

  • Implementations are documented in detail at docs.wacarbon.com, allowing users to see real-world data choices like hourly vs. monthly billing data.

  • The goal is to support adoption across different regulatory regimes by providing a common, transparent baseline measurement tool.

• Strategic intent behind developing these methodologies is to provide market participants and grid operators confidence that demand-side energy reductions are accurately measured and valued, potentially replacing less reliable baselines like the traditional 5-in-10 method (08:33)

  • This transparency aims to unlock the value of distributed energy resources (DERs) for load growth solutions.

  • It supports the emergence of virtual power plants and contractual settlements based on verified demand response performance.

Load Shifting Methodology

The load shifting methodology measures sustained energy consumption shifts from peak to off-peak periods using pre- and post-intervention baselines, focusing on persistent schedule changes rather than rare events.

• Load shifting is defined as ongoing energy use shifts from peak to off-peak periods, measured by comparing consumption before and after implementing new load schedules or optimizations (25:27)

  • The baseline period precedes the intervention date, similar to energy efficiency modeling.

  • Hourly consumption and local temperature data feed into the OpenDSM model to predict expected energy use.

  • Savings are calculated only for defined peak (energy reduction target) and off-peak (energy increase target) windows, which can be fixed or dynamic (e.g., day-ahead price or carbon signals).

• Savings accounting rules restrict counted load shifting to the lesser of increased off-peak or decreased peak energy within a rolling 24-hour window (32:21)

  • If off-peak energy increase exceeds peak reduction, only the peak reduction amount counts.

  • Unused off-peak “credits” expire after 24 hours to avoid indefinite carryover.

  • This ensures measured savings reflect real temporal shifts rather than net energy savings or storage.

• The methodology supports flexible window definitions that can vary daily based on signals such as grid price, carbon intensity, or curtailment forecasts (38:59)

  • Windows must be predefined independently of observed savings to preserve additionality and prevent gaming.

  • This adaptability enables alignment with various grid signals and market structures.

• Current limitations and future extensions include lack of synthetic baselines for new electrification loads and non-building load shifting scenarios like data center workload migration (31:22, 35:01)

  • Stephen Suffian acknowledged the need to develop methods for cases where no prior baseline data exists or where shifting occurs across locations.

  • Collaboration offers potential to extend methodologies to emerging use cases such as AI training workload shifts across data centers.

Carbon and Market Signal Integration

The methodologies are designed to integrate with diverse grid signals, including carbon intensity and dynamic pricing, enabling multi-attribute measurement of demand flexibility impacts.

• Savings can be expressed not only in energy units but also in carbon, dollars, and capacity values to support broad valuation needs (46:30)

  • Load shifting’s paired assignment of increased off-peak to decreased peak energy facilitates calculation of net carbon impact.

  • This supports strategies that optimize for emissions reduction in addition to cost savings.

• “Signals” defining peak and off-peak windows can originate from grid operators, utilities, or end users and include price, carbon intensity, or curtailment forecasts (49:26)

  • This flexibility allows methodologies to adapt to different market designs and customer goals.

  • Implementers choose prioritization frameworks, such as whether to prioritize load reduction or carbon reduction first, based on business or policy objectives.

• Challenges remain in accounting for stochastic, price-responsive loads such as EV charging or heat pump scheduling where consumption profiles vary unpredictably day-to-day (51:17)

  • Bruce Nordman highlighted that these “implicit” price responses complicate baseline modeling due to variable consumption patterns.

  • McGee Young suggested future methodology refinements will need to address how to model uncertainty and aggregate impacts in these cases.

Strategic Vision and Market Positioning

The Open EAC Alliance aims to establish transparent, open-source demand-side measurement standards to support evolving grid flexibility markets and virtual power plant contracts.

• The methodologies provide a foundation for credible settlements and contracting by allowing grid operators and DER owners to verify demand response and load shifting performance with consistent, transparent metrics (44:52)

  • This underpins the growth of virtual power plants and demand flexibility procurement.

  • It addresses the market need for confidence in the value delivered by increasingly important demand-side resources.

• The group is positioning these methodologies as open, community-driven standards rather than exclusive proprietary tools (10:48)

  • This openness encourages broad adoption and iterative improvement from diverse stakeholders.

  • It fosters a peer-reviewed ecosystem where different regulatory regimes and contract structures can apply the core approaches.

• Next steps include expanding methodologies to cover batteries, hybrid onsite generation, and carbon accounting complexities to capture a wider range of flexibility assets and use cases (55:30)

  • The team is actively researching how to track energy flows through batteries with first-in, first-out logic.

  • They aim to incorporate multi-asset interactions and hybrid grid/onsite power dynamics into future versions.

• Community engagement is ongoing via the OpenEAC Substack and GitHub, with quarterly meetings planned to update and refine methods (57:40)

  • The next meeting is scheduled for February 19th, 2024, with automatic invites for subscribers.

  • Contributors are encouraged to submit new methodologies or enhancements for collaborative development.

Process and Implementation Framework

The methodologies are supported by detailed Measurement & Verification (M&V) plans that specify data inputs, treatment of baseline periods, and operational choices to ensure clarity and replicability.

• M&V plans document specific implementation decisions such as data frequency (hourly vs. monthly), weather normalization approaches, and choices about baseline event inclusion (12:22)

  • This modular design allows methodologies to be adapted to specific regulatory or market contexts.

  • It provides transparency and auditability critical for settlement and verification purposes.

• The OpenDSM eemeter regression model forms the core modeling engine for both demand response and load shifting, ensuring consistency across methodologies (17:57)

  • It uses temperature and temporal patterns to create predictive baselines.

  • Shortened baselines for demand response enable event-specific evaluation without sacrificing model robustness.

• The alliance maintains version-controlled open repositories on GitHub to track methodology changes and foster collaborative improvements (12:22)

  • This process supports iterative refinement based on public feedback and empirical findings.

  • It ensures users can see historical changes and rationale behind updates.

• Operational constraints such as event rarity, baseline window length, and 24-hour assignment of load shifting savings are codified to maintain methodological integrity and avoid gaming (22:29, 32:21)

  • These guardrails balance simplicity, accuracy, and practical implementation challenges.

  • They also reduce noise and ensure measured impacts reflect real grid benefits.

• Ongoing dialogue with stakeholders is encouraged via blog comments and direct contact, supporting continuous feedback and methodology evolution (45:15)

  • The alliance seeks to incorporate diverse use cases and emerging technologies.

  • This engagement model supports community ownership and practical relevance.