Forecasting Residential Heating and Electricity Demand with Scalable, High-Resolution, Open-Source Models
Stephen J. Lee, Cailinn Drouin · 2025
Read Forecasting Residential Heating and Electricity Demand with Scalable, High-Resolution, Open-Source Models by Stephen J. Lee, Cailinn Drouin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-22873.txt.
Introduction
Stephen Lee and Cailinn Drouin forecast home energy use building by building in a 2025 paper. City planners and grid operators need to know which houses will call for heat and power, hour by hour. The standard research simulator sketches that demand only coarsely.
Their open models learn from real meter data in a mostly gas-heated region, so the electric patterns they master are lights, appliances, and cooling. Building footprints, heights, crowding, land use, and fine weather feed the predictions, which come with honest uncertainty bands. In back-tests, errors fell by a fifth to a quarter against the standard tool, and forecast quality jumped by more than half.
Catalog record
| CREATOR | Stephen J. Lee, Cailinn Drouin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-22873
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Forecasting Residential Heating and Electricity Demand with Scalable, High-Resolution, Open-Source Models Stephen J. Lee, Cailinn Drouin Abstract Electrifying space and water heating is a critical priority for the energy transition. The necessary widespread adoption of heat pumps will have significant impacts on the power grid. Studies report heating electrification may increase winter peak electricity demand by up to 70%, with some colder regions experiencing a more than fourfold increase in peak demand. Contending with this increased demand will necessitate unprecedented upgrades to the power grid. The process of upgrading the grid has a critical spatial dimension, as heating demand, electricity demand, and the capacity of existing grid infrastructure vary significantly across regions. Grid planning also involves a critical temporal dimension: short-term weather patterns and long-term climate change introduce complexities and uncertainties that can be difficult to quantify. However, most existing demand forecasts are provided and validated only at aggregated spatial scales, lack temporal detail, and provide single-valued predictions. Without accurate, probabilistic, and spatially and temporally resolved demand forecasts, planners risk misallocating scarce resources. We present a novel framework for high-resolution forecasting of residential heating and electricity demand using probabilistic deep learning models. We focus specifically on providing hourly building-level electricity and heating demand forecasts for the residential sector. Leveraging multimodal building-level information – including data on building footprint areas, heights, nearby building density, nearby building size, land use patterns, and high-resolution weather data – and probabilistic modeling, our methods provide granular insights into demand heterogeneity. Validation at the building level underscores a step change improvement in performance relative to NREL’s ResStock model, which has emerged as a research community standard for residential heating and electricity demand characterization. In building-level heating and electricity estimation backtests, our probabilistic models respectively achieve RMSE scores 18.3% and 35.1% lower than those based on ResStock. By offering an open-source, scalable, high-resolution platform for demand estimation and forecasting, this research advances the tools available for policymakers and grid planners, contributing to the broader effort to decarbonize the U.S. building stock and meeting climate objectives. 1 Introduction Electrifying space and water heating is a critical priority for the energy transition [ 1 , 2 ] . Residential and commercial buildings make up 13% of all U.S. emissions [ 3 ] , with fossil-fueled space heating representing the single greatest constituent of this share [ 4 ] . Heat pumps offer a promising solution to decarbonizing building heating. By enabling the efficient electrification of heating systems, they provide a pathway for significant emissions reductions as the power grid transitions to cleaner energy sources. The necessary widespread adoption of heat pumps will have significant impacts on the power grid. For instance, Waite and Modi estimate that full electrification of heating with current high-efficiency heat pump technologies could increase national peak electricity loads by 70%, with some colder regions experiencing a more than fourfold increase in peak demand [ 5 ] . Contending with this increased demand will necessitate unprecedented upgrades to the power grid. The process of upgrading the grid is inherently spatial in nature as heating demand, electricity demand, and the capacity of existing grid infrastructure varies widely across regions. Unfortunately, the vast majority of existing demand forecasts are provided and validated only at aggregated scales, obfuscating the heterogeneity and distribution of demand within communities. They also rarely provide forecasts accounting for short- and long-term trends. Without accurate, spatially and temporally resolved demand forecasts, planners risk critically misallocating scarce resources. Detailed demand forecasting is essential to address these challenges. By incorporating high levels of spatial and temporal granularity, advanced forecasting models can identify demand at arbitrary levels of spatial and temporal aggregation, estimate localized peak heating and electricity loads, anticipate seasonal variations, predict long-term trends, characterize forecasting uncertainty, and support grid planners to optimize asset investments. This level of detail helps to minimize costs, reduce inefficiencies, and enhance resilience as the grid adapts to support widespread electrification. We present a novel framework for high-resolution residential heating and electricity demand forecasting employing probabilistic deep learning models. We focus specifically on providing hourly building-level electricity and heating demand forecasts for the residential sector. Leveraging multimodal building-level information – including data on building footprints, heights, nearby building density, land use patterns, and high-resolution weather data – and probabilistic modeling, our methods provide granular insights into demand heterogeneity. Validation at the building level underscores a step change improvement in performance relative to NREL’s ResStock model, which has emerged as a research community standard for residential heating and electricity demand characterization. By offering an open-source, scalable, high-resolution platform for demand estimation and forecasting, this research advances the tools available for policymakers and grid planners, contributing to the broader effort to decarbonize the U.S. building stock and meet climate objectives. We make the following contributions: 1. We develop a scalable, data-driven framework for residential heating and electricity demand estimations and forecasting that provides building-level, hourly estimates. 2. Our models generate probabilistic rather than deterministic forecasts, offering more robust and reliable insights for grid planning and policy development. 3. We employ open and widely available remote sensing features, allowing us to provide scalable forecasts applicable across diverse geographic regions and timeframes. This enables historical analyses as well as climate scenario-based projections. 4. Our initial focus is in the U.S. Mid-Atlantic region, where our models demonstrate substantial improvements over industry-standard tools. Compared to NREL’s ResStock, our probabilsitic models achieves RMSE reductions of 18.3% for heating demand estimation and 35.1% for electricity demand estimation. These results underscore the enhanced accuracy and reliability of our approach for infrastructure planning. 5. By providing an open-source, scalable platform for demand estimation and forecasting, we facilitate accessibility for policymakers, researchers, and grid planners. 2 Related Work Various modeling approaches have been developed to estimate heating demand and assess the impacts of electrification, each with distinct advantages and limitations. We broadly categorize existing approaches into physics-based simulation models, spatially aggregated demand models, heating degree day regression models, and machine learning-based models. In this section, we review the state-of-the-art for each category, examine pros and cons, and contrast with our complimentary approach. 2.1 Physics-Based Simulation Models Physics-based simulation models estimate energy demand using detailed engineering principles and building physics. The National Renewable Energy Laboratory (NREL)’s ResStock and ComStock are widely used to generate end-use load profiles (EULPs) for the U.S. residential and commercial sectors, supporting energy planning and policy decisions [ 6 ] . These models construct representative samples of the U.S. building stock by correlating characteristics such as vintage, insulation levels, and HVAC efficiency using official census and survey data [ 7 , 8 ] . They simulate energy flows using the OpenStudio and EnergyPlus platforms, incorporating building attributes and weather conditions. Model outputs include hourly and sub-hourly energy consumption by fuel type and end use, calibrated against datasets such as the Residential Energy Consumption Survey (RECS), Commercial Buildings Energy Consumption Survey (CBECS), and energy consumption data at multiple resolutions. ResStock and ComStock rely on synthetic archetypes rather than actual building-level data, resulting in challenges when aligning model outputs with specific premises and limiting their accuracy for localized or peak demand estimation [ 6 , 9 ] . Their validation is performed at state or city levels, with aggregated accuracy metrics that