Thermal-GEMs: Generalized Models for Building Thermal Dynamics

Felix Koch; Fabian Raisch; Benjamin Tischler

eess.SYApr 7, 2026

Thermal-GEMs: Generalized Models for Building Thermal Dynamics

Authors: Felix Koch, Fabian Raisch, Benjamin Tischler

Organizations: Technical University of Applied Sciences Rosenheim, Rosenheim, Germany · Technical University of Munich, Munich, Germany

Abstract

Data-driven models for building thermal dynamics are a scalable approach for enabling energy-efficient operation through fault detection & diagnosis or advanced control. To obtain accurate models, measurement data from a target building spanning months to years are required. Transfer Learning (TL) mitigates this challenge by employing pretrained models based on single or multiple source buildings. General multi-source TL models promise to outperform single-source TL, but alternative multi-source modeling architectures remain to be explored, and evaluation on real-world data is missing. Moreover, time series foundation models (TSFM) have emerged as candidates for the best-performing general models. Hence, we conduct a first, comprehensive assessment of general modeling approaches for building thermal dynamics, including multi-source TL and TSFMs. Our assessment includes ablations using four state-of-the-art multi-source TL architectures and evaluations on synthetic as well as real-world data. We demonstrate that multi-source TL models are highly effective in accurately modeling buildings in real-world applications, yielding up to 63% lower forecasting errors compared to single-source TL. Moreover, our results suggest a trade-off between multi-source TL models exclusively pretrained with building data and TSFMs pretrained with a multitude of different time series, revealing that data from 16-32 source buildings must be available over 1 year for pretraining multi-source TL models to consistently outperform TSFMs as evaluated using the mean absolute error. These findings provide practical guidance for selecting modeling strategies based on the number of source buildings available for pretraining multi-source TL models.

Figures & tables

Figure 1. Overview of modeling strategies regarding our research questions.

Observable		Dataset
Name	Type	D1	D2	D3
Indoor Humidity	Feature	✓	$\times$	$\times$
Thermostat Heat Setpoint	Feature	✓	$\times$	✓
Thermostat Cool Setpoint	Feature	✓	$\times$	$\times$
Supply Air Fan Runtime	Feature	✓	$\times$	$\times$
Presence of Occupants	Feature	✓	$\times$	$\times$

Table 1. Features included in the three datasets.

	EcoBee	Ideal	Euro. House
# Rooms Total	932	227	450
# Sources Total	892	202	400
# Targets	40	25	50
% Targets	4.3	11.0	11.1

Table 2. Source and target split for all datasets

Figure 2. Seasonal Fine-Tuning split for the experiments

Figure 3. Zero-Shot MAE of all architectures as MSGMs

Figure 4. MAE of Single-Source TL versus MSGM for different fine-tunings on real-world data (Ecobee). MSGM beats single-source TL once errors are above the dotted line.

Figure 5. Zero-shot performance: TSFMs vs. the best-performing MSGMs (Transformer for Euro, LSTM for Ecobee and Ideal).

Figure 6. Zero-shot performance: Per-building TSFM compared to MSGM (each data point is a target building)

Figure 7. Zero-shot performance: Influence of the number of sources on the MSGM for buildings. The best-performing TSFM with respect to each dataset is used as a baseline.

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Hyperparameter	LSTM			Transformer			Mamba			xLSTM
Hyperparameter	From	To	Step	From	To	Step	From	To	Step	From	To	Step
Hidden Size	64	256	32	64	256	32	64	256	32	64	256	32
# Layers	2	4	1	2	4	1	2	8	1	2	4	1
Batch Size	32	128	32	32	128	32	32	128	32	32	128	32
Learning Rate	.0001	.001	.0001	.0001	.001	.0001	.0001	.001	.0001	.0001	.001	.0001
Dropout	0.0	0.1	0.01	0.0	0.1	0.01	-	-	-	0.0	0.1	0.01

Appendix

Table 3. Hyperparameter Search Space for the Multi-Source General Models.

Parameter	LSTM			Transformer			Mamba			xLSTM
Parameter	Ecobee	Ideal	Euro	Ecobee	Ideal	Euro	Ecobee	Ideal	Euro	Ecobee	Ideal	Euro
Hidden Size	96	128	192	192	128	192	160	128	96	96	128	64
# Layers	3	3	3	4	3	3	3	4	3	3	3	3
Batch Size	128	128	96	128	128	128	128	128	96	128	128	128
Use AMP	True	True	True	True	True	True	True	True	True	False	False	False
Learning Rate	.0009	.001	.001	.0001	.0005	.0004	.0006	.0002	.0002	.0002	.0005	.0005

Appendix

Table 4. Hyperparameters for all MSGM architectures for each dataset.

Figure 8. RMSE of Single-Source TL versus MSGM for different fine-tunings on real-world data (Ecobee). MSGM beats single-source TL once errors are above the dotted line.

Figure 9. Zero-shot performance RMSE: Influence of the number of sources on the MSGM for buildings. The best-performing TSFM with respect to each dataset is used as a baseline.

Time Series Forecasting

Time series forecasting predicts future values from ordered observations using statistical, neural, and probabilistic models to support decisions under uncertainty.

See all 858 papers

GenTL: A General Transfer Learning Model for Building Thermal Dynamics
Jan 23, 2025Fabian Raisch, Thomas Krug, Christoph Goebel +1Fine-Tuning Transfer Learning
Technical University of Applied Sciences Rosenheim Rosenheim, Germany · Technical University of Munich Munich, Germany
Toward a foundational thermal model for residential buildings
May 2, 2026Ting-Yu Dai, Kingsley Nweye, Dev Niyogi +1Time Series Forecasting Physics-Informed ML
The University of Texas at Austin · Eindhoven University of Technology
Very Exciting: Zero-Shot Model Predictive Control of Buildings via Excitation-Based Generalized Transfer Learning Models
Sep 11, 2026Fabian Raisch, Felix Koch, Zack Xuereb Conti +2Building Energy Management Model Predictive Control
Technical University of Applied Sciences Rosenheim, Hochschulstraße 1, Rosenheim, 83024, Germany · Technical University of Munich, Arcisstraße 21, Munich, 80333, Germany · Karlsruhe Institute of Technology, Hermann-von-Helmholtz-Platz 1, Eggenstein-Leopoldshafen, 76344, Germany +1

Jan 23, 2025J/K move · Enter open · S save

GenTL: A General Transfer Learning Model for Building Thermal Dynamics

Fabian Raisch, Thomas Krug, Christoph Goebel, Benjamin Tischler

Technical University of Applied Sciences Rosenheim Rosenheim, Germany · Technical University of Munich Munich, Germany

Fine-Tuning Transfer Learning

Open paper AskPDF arXiv

Transfer Learning (TL) is an emerging field in modeling building thermal dynamics. This method reduces the data required for a data-driven model of a target building by leveraging knowledge from a source building. Consequently, it enables the creation of data-efficient models that can be used for advanced control and fault detection & diagnosis. A major limitation of the TL approach is its inconsistent performance across different sources. Although accurate source-building selection for a target is crucial, it remains a persistent challenge. We present GenTL, a general transfer learning model for single-family houses in Central Europe. GenTL can be efficiently fine-tuned to a large variety of target buildings. It is pretrained on a Long Short-Term Memory (LSTM) network with data from 450 different buildings. The general transfer learning model eliminates the need for source-building selection by serving as a universal source for fine-tuning. Comparative analysis with conventional single-source to single-target TL demonstrates the efficacy and reliability of the general pretraining approach. Testing GenTL on 144 target buildings for fine-tuning reveals an average prediction error (RMSE) reduction of 42.1 % compared to fine-tuning single-source models.

Abstract

Time Series Forecasting

Explore similar work

GenTL: A General Transfer Learning Model for Building Thermal Dynamics

Toward a foundational thermal model for residential buildings

Very Exciting: Zero-Shot Model Predictive Control of Buildings via Excitation-Based Generalized Transfer Learning Models