Toward a foundational thermal model for residential buildings

• URL: http://arxiv.org/abs/2605.01364v1
• Date: Sat, 02 May 2026 10:19:24 GMT
• Title: Toward a foundational thermal model for residential buildings
• Abstract summary: We present a physics-informed transformer architecture that embeds domain knowledge into a decoder-only framework.<n>We evaluate our model on the CityLearn dataset spanning 247 residential buildings across three climate zones.<n>Our results establish physics-informed architectural principles as a promising foundation for universal building thermal models.
• Score: 1.628785836948598
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The building energy community lacks a foundational thermal model, i.e., a single pretrained model capable of generalizing across diverse buildings, climates, and control strategies without building-specific calibration. Achieving this vision requires architectural principles that capture universal thermal dynamics rather than memorizing building-specific patterns. We take a step toward this goal by presenting a physics-informed transformer architecture that embeds domain knowledge, e.g., derivative enrichment and Euler-based numerical integration, into a decoder-only framework. We incorporate static building features extracted from simulation models and employ Rotary Position Embedding attention to capture temporal dependencies. Evaluated on the CityLearn dataset spanning 247 residential buildings across three climate zones, our model achieves one-step prediction accuracy (RMSE of 0.30°C in Texas, 0.29°C in Vermont) while outperforming both traditional baselines and fine-tuned Time-Series Foundation Models. We also demonstrate zero-shot transferability: models trained on as few as two buildings generalize to unseen buildings and climate zones without fine-tuning. Despite the limitation of simulated residential buildings, our results establish physics-informed architectural principles as a promising foundation for universal building thermal models.

Related papers

• ArchEGraph: A Large-Scale Graph Dataset for Geometry-Topology-Physics Aligned Building Energy Modeling [13.018187478378422]
  ArchEGraph is a large-scale benchmark dataset that represents buildings as heterogeneous graphs with aligned geometry, topology, weather, and zone-level thermal loads.<n>The dataset contains 5,481 buildings and 49,326 validated building-weather simulation cases.
  arXiv  Detail & Related papers  (2026-08-07T03:45:41Z)
• Lights, Camera, Carbon: Architectural Scaling Laws for Video Generation Energy Consumption [4.930883100315264]
  We present a framework for estimating the energy consumption of text-to-video (T2V) and text-to-video-audio (T2VA) models.<n> Forward, it predicts energy from generation parameters and architectural principles; backward, it recovers architectural scaling behavior from observed inference times.<n>This approach offers a standardized, empirically and theoretically grounded framework for sustainability benchmarking across T2V models and architectures.
  arXiv  Detail & Related papers  (2026-07-05T23:58:29Z)
• Earth-o1: A Grid-free Observation-native Atmospheric World Model [81.55899748753434]
  We present Earth-o1, an observation-native atmospheric world model that overcomes structural limitations.<n>By integrating diverse sensor inputs into a unified, grid-free dynamical field, the model autonomously advances the atmospheric state in space and time.<n>We show that this fundamentally distinct paradigm enables direct, real-time forecasting and cross-sensor inference without the overhead of explicit numerical solvers.
  arXiv  Detail & Related papers  (2026-05-07T14:27:48Z)
• HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents [71.09493646932046]
  We introduce HY-Embodied-0.5, a family of foundation models specifically designed for real-world embodied agents.<n>The HY-Embodied-0.5 suite comprises two primary variants: an efficient model with 2B activated parameters designed for edge deployment, and a powerful model with 32B activated parameters targeted for complex reasoning.<n>Our MoT-2B model outperforms similarly sized state-of-the-art models on 16 benchmarks, while the 32B variant achieves performance comparable to frontier models such as Gemini 3.0 Pro.
  arXiv  Detail & Related papers  (2026-04-08T17:59:48Z)
• The Trinity of Consistency as a Defining Principle for General World Models [106.16462830681452]
  General World Models are capable of learning, simulating, and reasoning about objective physical laws.<n>We propose a principled theoretical framework that defines the essential properties requisite for a General World Model.<n>Our work establishes a principled pathway toward general world models, clarifying both the limitations of current systems and the architectural requirements for future progress.
  arXiv  Detail & Related papers  (2026-02-26T16:15:55Z)
• Assessing Electricity Demand Forecasting with Exogenous Data in Time Series Foundation Models [0.0]
  This paper empirically evaluates foundation models capable of modeling cross-channel correlations against a baseline LSTM.<n>We find that the simple baseline frequently outperforms all foundation models in Singapore's stable climate, particularly for short-term horizons.<n>These results challenge assumptions about universal foundation model superiority and highlight the need for domain-specific models.
  arXiv  Detail & Related papers  (2026-02-05T07:17:21Z)
• BuildingWorld: A Structured 3D Building Dataset for Urban Foundation Models [17.325315633493624]
  BuildingWorld is a globally representative dataset for urban-scale foundation modeling and analysis.<n>It provides about five million LOD2 building models collected from diverse sources, accompanied by real and simulated airborne LiDAR point clouds.<n>Cyber City is a virtual city model to enable the generation of unlimited training data with customized and structurally diverse point cloud distributions.
  arXiv  Detail & Related papers  (2025-11-09T11:40:34Z)
• Aether: Geometric-Aware Unified World Modeling [49.33579903601599]
  Aether is a unified framework that enables geometry-aware reasoning in world models.<n>Our framework achieves zero-shot generalization in both action following and reconstruction tasks.<n>We hope our work inspires the community to explore new frontiers in physically-reasonable world modeling.
  arXiv  Detail & Related papers  (2025-03-24T17:59:51Z)
• Global Transformer Architecture for Indoor Room Temperature Forecasting [49.32130498861987]
  This work presents a global Transformer architecture for indoor temperature forecasting in multi-room buildings. It aims at optimizing energy consumption and reducing greenhouse gas emissions associated with HVAC systems. Notably, this study is the first to apply a Transformer architecture for indoor temperature forecasting in multi-room buildings.
  arXiv  Detail & Related papers  (2023-10-31T14:09:32Z)
• BuildingsBench: A Large-Scale Dataset of 900K Buildings and Benchmark for Short-Term Load Forecasting [3.8489470606363514]
  Data-driven short-term load forecasting (STLF) has suffered from a lack of open, large-scale datasets with high building diversity. We present BuildingsBench, which consists of: 1) Buildings-900K, a large-scale dataset of 900K simulated buildings representing the U.S. building stock; and 2) an evaluation platform with over 1,900 real residential and commercial buildings from 7 open datasets.
  arXiv  Detail & Related papers  (2023-06-30T21:26:24Z)
• A physics-based domain adaptation framework for modelling and forecasting building energy systems [5.8010446129208155]
  State-of-the-art machine-learning-based models are a popular choice for modeling and forecasting energy behavior in buildings. However, their architecture does not hold physical to mechanistic structures linked with governing physical phenomena. We introduce a novel SDA approach where instead of labeled data, we leverage the geometric structure of the LTI governed by heat transfer ordinary differential equations.
  arXiv  Detail & Related papers  (2022-08-19T17:27:39Z)
• Low Emission Building Control with Zero-Shot Reinforcement Learning [70.70479436076238]
  Control via Reinforcement Learning (RL) has been shown to significantly improve building energy efficiency. We show it is possible to obtain emission-reducing policies without a priori--a paradigm we call zero-shot building control.
  arXiv  Detail & Related papers  (2022-08-12T17:13:25Z)
• Physics-constrained Deep Learning of Multi-zone Building Thermal Dynamics [6.970705794911342]
  We present a physics-constrained control-oriented deep learning method for modeling building thermal dynamics. Specifically, our method incorporates structural priors from traditional physics-based building modeling into the neural network thermal dynamics model structure. We demonstrate the proposed data-driven modeling approach's effectiveness and physical interpretability on a dataset obtained from a real-world office building with 20 thermal zones.
  arXiv  Detail & Related papers  (2020-11-11T06:39:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.