Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data driven modeling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,027 records · Page 57
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Validation of theory-based models for the control of plasma currents in W7-X divertor plasmas

A theory-based model for the control of plasma currents for steady-state operation in W7-X is proposed and intended for model-based plasma control. The conceptual outline implies the strength of physics-based models: it offer approaches applicable to future conditions of fusion devices or next-step machines. The application at extrapolated settings is related to the validity range of the theory model. Therefore, the predictive power of theory-based control models could be larger than for data-driven approaches and limitations can be predicted from the validity range for the prediction of bootstrap currents in W7-X. The model predicts the L/R response when density or heating power is changed. The model is based on neoclassical bootstrap current calculations and validated for different discharge conditions. While the model was found to be broadly applicable for conducted electron-cyclotron-heated discharges in W7-X, limits were found for cases when the polarization of the electron cyclotron heating was changed from X2 to O2-heating. The validity assessment attempts to quantify the potential of the derived model for model-based control in the operational space (density, heating power) of W7-X.

neoclassical modelling↗
DOE OSTI2024

Determining Stellar Elemental Abundances from DESI Spectra with the Data-driven Payne

Abstract Stellar abundances for a large number of stars provide key information for the study of Galactic formation history. Large spectroscopic surveys such as the Dark Energy Spectroscopic Instrument (DESI) and LAMOST take median-to-low-resolution (R≲ 5000) spectra in the full optical wavelength range for millions of stars. However, the line-blending effect in these spectra causes great challenges for elemental abundance determination. Here we employDD-Payne, a data-driven method regularized by differential spectra from stellar physical models, to the DESI early data release spectra for stellar abundance determination. Our implementation delivers 15 labels, including effective temperatureT eff , surface gravity log g , microturbulence velocityv mic , and the abundances for 12 individual elements, namely C, N, O, Mg, Al, Si, Ca, Ti, Cr, Mn, Fe, and Ni. Given a spectral signal-to-noise ratio of 100 per pixel, the internal precisions of the label estimates are about 20 K forT eff , 0.05 dex for log g , and 0.05 dex for most elemental abundances. These results agree with the theoretical limits from the Crámer–Rao bound calculation within a factor of 2. The majority of the accreted halo stars contributed by the Gaia–Enceladus–Sausage are discernible from the disk and in situ halo populations in the resultant [Mg/Fe]–[Fe/H] and [Al/Fe]–[Fe/H] abundance spaces. We also provide distance and orbital parameters for the sample stars, which spread over a distance out to ∼100 kpc. The DESI sample has a significantly higher fraction of distant (or metal-poor) stars than the other existing spectroscopic surveys, making it a powerful data set for studying the Galactic outskirts. The catalog is publicly available.

Astronomy & Astrophysics↗
DOE OSTI2024

TNet: A Model-Constrained Tikhonov Network Approach for Inverse Problems

Deep learning (DL), in particular deep neural networks, by default is purely data-driven and in general does not require physics. This is the strength of DL but also one of its key limitations when applied to science and engineering problems in which underlying physical properties—such as stability, conservation, and positivity—and accuracy are required. DL methods in their original forms are often not capable of respecting the underlying mathematical models or achieving desired accuracy even in big-data regimes. On the other hand, many data-driven science and engineering problems, such as inverse problems, typically have limited experimental or observational data, and DL would overfit the data in this case. Leveraging information encoded in the underlying mathematical models, we argue, not only compensates for missing information in low data regimes but also provides opportunities to equip DL methods with the underlying physics, hence promoting better generalization. This paper develops a model-constrained DL approach and its variant TNet—a Tikhonov neural network—which are capable of learning not only information hidden in the training data but also in the underlying mathematical models to solve inverse problems governed by partial differential equations in low data regimes. We provide the constructions and some theoretical results for the proposed approaches for both linear and nonlinear inverse problems. Since TNet is designed to learn inverse solutions with Tikhonov regularization, it is interpretable: in fact it recovers Tikhonov solutions for linear cases while potentially approximating Tikhonov solutions for nonlinear inverse problems. We also prove that data randomization can enhance not only the smoothness of the networks but also their generalizations. Comprehensive numerical results confirm the theoretical findings and show that with even as little as 1 training data sample for one-dimensional (1D) deconvolution, 5 for an inverse 2D heat conductivity problem, 100 for inverse initial conditions for a time-dependent 2D Burgers’s equation, and 50 for inverse initial conditions for 2D Navier–Stokes equations, TNet solutions can be as accurate as Tikhonov solutions while being several orders of magnitude faster. Furthermore, this is possible owing to the model-constrained term, replications, and randomization.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Improving neutrino-nuclei interaction models: Recommendations and case studies on Peelle’s Pertinent Puzzle

Improving the modeling of neutrino-nuclei interactions using data-driven methods is crucial for high-precision neutrino oscillation experiments. This paper investigates Peelle’s Pertinent Puzzle (PPP) in the context of neutrino measurements, a longstanding challenge to fitting theoretical models to experimental data. Inconsistencies in data-model comparisons hinder efforts to enhance the accuracy and reliability of model predictions. We analyze various sources contributing to these inconsistencies and propose strategies to address them, supported by practical case studies. We advocate for incorporating model fitting exercises as a standard practice in cross section publications to enhance the robustness of results. We use a common analysis framework to explore PPP-related challenges with MicroBooNE and T2K data in an unified manner. Our findings offer valuable insights for improving the accuracy and reliability of neutrino-nuclei interaction models, particularly by systematically tuning models using data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

OpenET: Filling a Critical Data Gap in Water Management for the Western United States

The lack of consistent, accurate information on evapotranspiration (ET) and consumptive use of water by irrigated agriculture is one of the most important data gaps for water managers in the western United States (U.S.) and other arid agricultural regions globally. The ability to easily access information on ET is central to improving water budgets across the West, advancing the use of data-driven irrigation management strategies, and expanding incentive-driven conservation programs. Recent advances in remote sensing of ET have led to the development of multiple approaches for field-scale ET mapping that have been used for local and regional water resource management applications by U.S. state and federal agencies. The OpenET project is a community-driven effort that is building upon these advances to develop an operational system for generating and distributing ET data at a field scale using an ensemble of six well-established satellite-based approaches for mapping ET. Key objectives of OpenET include: Increasing access to remotely sensed ET data through a web-based data explorer and data services; supporting the use of ET data for a range of water resource management applications; and development of use cases and training resources for agricultural producers and water resource managers. Here we describe the OpenET framework, including the models used in the ensemble, the satellite, meteorological, and ancillary data inputs to the system, and the OpenET data visualization and access tools. We also summarize an extensive intercomparison and accuracy assessment conducted using ground measurements of ET from 139 flux tower sites instrumented with open path eddy covariance systems. Results calculated for 24 cropland sites from Phase I of the intercomparison and accuracy assessment demonstrate strong agreement between the satellite-driven ET models and the flux tower ET data. For the six models that have been evaluated to date (ALEXI/DisALEXI, eeMETRIC, geeSEBAL, PT-JPL, SIMS, and SSEBop) and the ensemble mean, the weighted average mean absolute error (MAE) values across all sites range from 13.6 to 21.6 mm/month at a monthly timestep, and 0.74 to 1.07 mm/day at a daily timestep. At seasonal time scales, for all but one of the models the weighted mean total ET is within ±8% of both the ensemble mean and the weighted mean total ET calculated from the flux tower data. Overall, the ensemble mean performs as well as any individual model across nearly all accuracy statistics for croplands, though some individual models may perform better for specific sites and regions. We conclude with three brief use cases to illustrate current applications and benefits of increased access to ET data, and discuss key lessons learned from the development of OpenET.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

High-throughput solubility determination for data-driven materials design and discovery in redox flow battery research

Solubility is crucial for redox flow batteries because it affects their energy density. A data-driven approach based on artificial intelligence/machine learning models can accelerate the development of highly soluble redox-active materials, but the lack of relevant, large-quantity data makes accurate solubility prediction difficult. To overcome this deficiency, we developed a high-throughput experimentation process that combines a robotically controlled platform with high-throughput methodology to collect large-scale and high-quality solubility data. We demonstrate the potential utility and applicability of this high-throughput process by measuring the aqueous and non-aqueous solubilities of redox-active materials and studying the effect of additives on their solubilities for both aqueous and non-aqueous redox flow battery applications. A redox flow battery based on our optimized negative electrolyte formulation and a ferrocyanide-positive electrolyte offers highly stable performance over 18 days (>100 cycles) with consistent capacity and a 24% boost in energy density.

25 ENERGY STORAGE↗
DOE OSTI2024

Learning thermodynamically constrained equations of state with uncertainty

Numerical simulations of high energy-density experiments require equation of state (EOS) models that relate a material’s thermodynamic state variables—specifically pressure, volume/density, energy, and temperature. EOS models are typically constructed using a semi-empirical parametric methodology, which assumes a physics-informed functional form with many tunable parameters calibrated using experimental/simulation data. Since there are inherent uncertainties in the calibration data (parametric uncertainty) and the assumed functional EOS form (model uncertainty), it is essential to perform uncertainty quantification (UQ) to improve confidence in EOS predictions. Model uncertainty is challenging for UQ studies since it requires exploring the space of all possible physically consistent functional forms. Thus, it is often neglected in favor of parametric uncertainty, which is easier to quantify without violating thermodynamic laws. This work presents a data-driven machine learning approach to constructing EOS models that naturally captures model uncertainty while satisfying the necessary thermodynamic consistency and stability constraints. We propose a novel framework based on physics-informed Gaussian process regression (GPR) that automatically captures total uncertainty in the EOS and can be jointly trained on both simulation and experimental data sources. A GPR model for the shock Hugoniot is derived, and its uncertainties are quantified using the proposed framework. We apply the proposed model to learn the EOS for the diamond solid state of carbon using both density functional theory data and experimental shock Hugoniot data to train the model and show that the prediction uncertainty is reduced by considering thermodynamic constraints.

Sharma, Himanshu (ORCID:000900050235718X)↗
DOE OSTI2025

Unified 0.25-degree gridded infrastructure-critical extreme weather for the United States from 1979 to 2100

Extreme weather events can severely disrupt critical infrastructure, triggering cascading effects on power, transportation, and essential services. However, standard weather and climate datasets often lack specialized variables necessary for hazard assessments. We present a unified dataset of infrastructure-critical weather and climate variables across the United States at 0.25° resolution, covering daily or sub-daily intervals from 1979 to 2100. The dataset includes temperature, dew point, wind gusts, precipitation partitioned by rain, snow, and freezing rain or ice pellets, lightning, and wildfire metrics. Historical conditions (1979-2023) are synthesized from observations and reanalysis products, while future projections are derived from 14 CMIP6 global climate models (historical, SSP245, and SSP585 experiments). Physically based and data-driven methods are used to estimate variables not directly provided by existing models. By integrating these variables into a single unified dataset, we enable consistent, high-resolution assessments of weather-related infrastructure risks across past and future periods, supporting wide-ranging applications in energy, transportation, water resources, emergency management, and beyond.

Climate and Earth system modelling↗
DOE OSTI2026

Latent Twins

Over the past decade, scientific machine learning has transformed the development of mathematical and computational frameworks for analyzing, modeling, and predicting complex systems. From inverse problems to numerical partial differential equations (PDEs), dynamical systems, and model reduction, these advances have pushed the boundaries of what can be simulated. Yet they have often progressed in parallel, with representation learning and algorithmic solution methods evolving largely as separate pipelines. With Latent Twins, we propose a unifying mathematical framework that creates a hidden surrogate in latent space for the underlying equations. Whereas digital twins mirror physical systems in the digital world, Latent Twins mirror mathematical systems in a learned latent space governed by operators. Through this lens, classical modeling, inversion, model reduction, and operator approximation all emerge as special cases of a single principle. We establish the fundamental approximation properties of Latent Twins for both ordinary differential equations (ODEs) and PDEs and demonstrate the framework across three representative settings: (i) canonical ODEs, capturing diverse dynamical regimes; (ii) a PDE benchmark using the shallow-water equations, contrasting Latent Twin simulations with deep operator network and forecasts with a four-dimensional variational method baseline; and (iii) a challenging real-data geopotential reanalysis dataset, reconstructing and forecasting from sparse, noisy observations. Latent Twins provide a compact, interpretable surrogate for solution operators that evaluate across arbitrary time gaps in a single-shot, while remaining compatible with scientific pipelines such as assimilation, control, and uncertainty quantification. Looking forward, this framework offers scalable, theory-grounded surrogates that bridge data-driven representation learning and classical scientific modeling across disciplines.

Latent Twins↗
DOE OSTI2023

Predicting future well performance for environmental remediation design using deep learning

Here in this study, we developed a deep learning (DL) framework with a multi-channel three-dimensional convolutional neural network (MC3D-CNN) to predict well performance and thereby assist future environmental remediation design. Such prediction of extraction well performance at designated locations is critical for configuring pump-and-treat (P&T) well network design and operation, setting reasonable target closure dates for overall remedying, and estimating remedy costs. The framework is developed with operational and monitoring data routinely collected during P&T remedy operations, including well extraction and injection rates as well as in situ contaminant concentrations. Traditionally, the collected data were rarely used for purposes other than assessing past well performance and the accuracy of the conceptual site model. However, recent advances in data-driven computational approaches enable better use of the large datasets to inform future well performance, enhance site characterization, and improve remediation planning. In this study, we established a DL framework to integrate transient three-dimensional contaminant plumes and multiple aquifer properties (e.g., hydraulic conductivity and hydrostratigraphic maps) to identify characteristic patterns controlling and representing extraction well mass recovery, aiming at providing future mass recovery estimates for existing wells and candidate wells at any proposed locations. We evaluated our framework by using a realistic synthetic dataset generated from a well-calibrated flow and transport model used in the 200 West Area of the U.S. Department of Energy’s Hanford Site in southeastern Washington state. The multi-channel feature in our framework allows integration of various types and temporal densities of training datasets for DL model development. Overall, we found that the trained DL model achieved an accuracy of over 90% in ranking extraction well performance in validation datasets, and over 80% in predicting high-performance-ranking well locations. This data-informed approach provides a flexible tool to support adaptive site management, streamline decision-making, and potentially reduce remediation time and costs. Our DL framework can be used as a filtering tool to improve the current P&T network optimization design by reducing the number of candidate well locations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Advancing Urban Water Resilience: Coproducing Knowledge through Civic–Academic Global Partnerships on Water and Climate

As extreme weather events become more pronounced, the vulnerabilities associated with the urban water supply and wastewater systems in megacities are intensified in multiple interconnected dimensions. These multifaceted water challenges can benefit from enhanced cross-sectoral collaboration and sharing of critical knowledge, which are essential for sustainable and adaptive water governance frameworks. In this context, the Megacity Alliance for Water and Climate (MAWAC)–Europe and North America Region (ENAR) Working Group convened a workshop in March 2023, followed by a subsequent workshop in London, United Kingdom, from 11 to 13 September 2024. These workshops aimed to investigate and devise solutions for the cascading hazards with water systems. The solutions examined various aspects focused on climate adaptation and mitigation, stormwater management, and the governance of water and wastewater systems. Additionally, discussions highlighted the importance of community engagement, economic considerations, equity, and effective communication in addressing these pressing challenges. Over the course of 3 days, experts from academia, government agencies, and industry engaged in meaningful discussions on digital modeling for integrated water management, climate-informed urban planning, and public–private–academic partnerships (Fig. 1). Case studies from cities such as New York, Los Angeles, London, Paris, and Chicago highlighted innovative governance strategies for managing water and wastewater systems, promoting water reuse, planning infrastructure, and fostering stakeholder-driven and stakeholder-informed adaptation. The workshop participants emphasized the need for data-driven decision-making, scalable governance models, and knowledge-sharing networks to enhance urban water governance for sustainability and resilience. This workshop report presents the key takeaways from the 3-day convening, providing a roadmap for integrating scientific research, policy frameworks, and emerging technologies to address water challenges faced by megacities.

Hydrologic models↗
DOE OSTI2022

Data-Driven Discovery of Active Nematic Hydrodynamics

Active nematics can be modeled using phenomenological continuum theories that account for the dynamics of the nematic director and fluid velocity through partial differential equations (PDEs). While these models provide a statistical description of the experiments, the relevant terms in the PDEs and their parameters are usually identified indirectly. We adapt a recently developed method to automatically identify optimal continuum models for active nematics directly from spatio-temporal data, via sparse regression of the coarse-grained fields onto generic low order PDEs. After extensive benchmarking, we apply the method to experiments with microtubule-based active nematics, finding a surprisingly minimal description of the system. Furthermore, our approach can be generalized to gain insights into active gels, microswimmers, and diverse other experimental active matter systems.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Data-driven Cyberattack Detection for Photovoltaic (PV) Systems through Analyzing Micro-PMU Data

With increasing exposure to software-based sensing and control, Photovoltaic (PV) systems are facing higher risks of cyber attacks. Here, to ensure the system stability and minimize potential economic losses, it is imperative to monitor operating states and detect attacks at the early stage. To meet this demand, Micro-Phasor Measurement Units (μPMU) are increasingly popular in monitoring distribution networks. However, due to the relatively low sampling rate, μPMU has not yet been used to detect and classify cyber-attacks in power electronics enabled smart grid. To our knowledge, this is one of the first attempts to use μPMU to detect cyber attacks that degrade the performance of power electronics systems. We propose to apply data-driven methods on micro-PMU data to implement attack detection. We have evaluated data-driven methods, including decision tree (DT), K-nearest neighbor (KNN), support vector machine (SVM), artificial neural network (ANN), long short-term memory (LSTM) and convolutional neural network (CNN). The proposed CNN model achieves the required performances with the highest 99.23% accuracy and 0.9963 F 1 score.

14 SOLAR ENERGY↗
DOE OSTI2025

Search for $t\bar{t}H/A \rightarrow t\bar{t}t\bar{t}$ production in proton–proton collisions at $\sqrt{s}=13$ $\text {TeV}$ with the ATLAS detector

A search is presented for a heavy scalar (H) or pseudo-scalar (A) predicted by the two-Higgs-doublet models, where the H/A is produced in association with a top-quark pair $(t\bar{t}H/A),$ and with the H/A decaying into a $t\bar{t}$ pair. The full LHC Run 2 proton–proton collision data collected by the ATLAS experiment is used, corresponding to an integrated luminosity of $139~\text {fb}^{-1}.$ Events are selected requiring exactly one or two opposite-charge electrons or muons. Data-driven corrections are applied to improve the modelling of the $t\bar{t}$ +jets background in the regime with high jet and b-jet multiplicities. These include a novel multi-dimensional kinematic reweighting based on a neural network trained using data and simulations. An H/A-mass parameterised graph neural network is trained to optimise the signal-to-background discrimination. In combination with the previous search performed by the ATLAS Collaboration in the multilepton final state, the observed upper limits on the $t\bar{t}H/A \rightarrow t\bar{t}t\bar{t}$ production cross-section at 95% confidence level range between 14 fb and 5.0 fb for an H/A with mass between 400 GeV and 1000 GeV , respectively. Assuming that both the H and A contribute to the $t\bar{t}t\bar{t}$ cross-section, tan β values below 1.7 or 0.7 are excluded for a mass of 400 GeV or 1000 GeV , respectively. The results are also used to constrain a model predicting the pair production of a colour-octet scalar, with the scalar decaying into a $t\bar{t}$ pair.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Classification of Nuclear Reactor Operations Using Spatial Importance and Multisensor Networks

Distributed multisensor networks record multiple data streams that can be used as inputs to machine learning models designed to classify operations relevant to proliferation at nuclear reactors. The goal of this work is to demonstrate methods to assess the importance of each node (a single multisensor) and region (a group of proximate multisensors) to machine learning model performance in a reactor monitoring scenario. This, in turn, provides insight into model behavior, a critical requirement of data-driven applications in nuclear security. Using data collected at the High Flux Isotope Reactor at Oak Ridge National Laboratory via a network of Merlyn multisensors, two different models were trained to classify the reactor’s operational state: a hidden Markov model (HMM), which is simpler and more transparent, and a feed-forward neural network, which is less inherently interpretable. Traditional wrapper methods for feature importance were extended to identify nodes and regions in the multisensor network with strong positive and negative impacts on the classification problem. These spatial-importance algorithms were evaluated on the two different classifiers. The classification accuracy was then improved relative to baseline models via feature selection from 0.583 to 0.839 and from 0.811 ± 0.005 to 0.884 ± 0.004 for the HMM and feed-forward neural network, respectively. While some differences in node and region importance were observed when using different classifiers and wrapper methods, the nodes near the facility’s cooling tower were consistently identified as important—a conclusion further supported by studies on feature importance in decision trees. Node and region importance methods are model-agnostic, inform feature selection for improved model performance, and can provide insight into opaque classification models in the nuclear security domain.

Tibbetts, Jake↗
DOE OSTI2021

EPIsembleVis: A geo-visual analysis and comparison of the prediction ensembles of multiple COVID-19 models

In this work, we present EPIsembleVis, a web-based comparative visual analysis tool for evaluating the consistency of multiple COVID-19 prediction models. Our approach analyzes a collection of COVID-19 predictions from different epidemiological models as an ensemble and utilizes two metrics to quantify model performance. These metrics include (a) prediction uncertainty (represented as the dispersion of predictions in each ensemble) and (b) prediction error (calculated by comparing individual model predictions with the recorded data). Through an interactive visual interface, our approach provides a data-driven workflow for (a) selecting and constructing the COVID-19 model prediction ensemble based on the spatiotemporal overlap of available predictions of multiple epidemiological models, (b) quantifying the model performance using both the uncertainty of each model prediction ensemble, and the error of each ensemble member that represents individual model predictions, and (c) visualizing the spatiotemporal variability in the projection performance of individual models using a suite of novel ensemble visualization techniques, such as the data availability map, a spatiotemporal textured-tile calendar, multivariate rose chart, and time-series leaflet glyph. We demonstrate the capability of our ensemble visual interface through a case study that investigates the performance of weekly COVID-19 predictions, which are provided through the COVID-19 Forecast Hub UMass-Amherst Influenza Forecasting Center of Excellence [47] for the United States and United States Territories. The EPIsembleVis tool is implemented using open-source web technologies and adaptive system design, rendering it interoperable with Elasticsearch and Kibana for automatically ingesting COVID-19 predictions from online repositories, and it is generalizable for analyzing worldwide projections from more epidemiological models.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

DINGO: Digital assistant to grid operators for resilience management of power distribution system

With increasing adverse weather events and disasters, enabling resiliency of the power distribution system (PDS) is becoming increasingly important. Here in this work, resiliency is defined as the systems ability to keep supplying critical loads even with multiple contingencies. Resiliency may depend on: (a) advanced tools to assist operators in situational awareness and decision making with the increasing volume of data generated by the PDS, (b) visualization and ease of interaction with system resources and information, especially during extreme events and resulting human operator stress, and (c) flexible resources and autonomous control. Operators and support engineers need to interact with the system for key information and take action under stress, given the requirement for decisions in a short time. Integrated technological solutions are prevailing steps to support the most appropriate decision during critical times to serve essential loads. In order to meet the required goals, a Real-time Resiliency Monitoring and Operational Decision Support (RT-RMOD) tool have been developed. It supports various functionalities, including real-time monitoring, resilience assessment, and proactive decision support. However, this work makes advanced feature additions to the tool by developing data-enabled resilience management algorithms for (i) outage detection and localization, (ii) Resiliency-metric driven restoration and reconfiguration, and (iii) NLP based digital assistant for operators called DINGO (DIgital assistaNt to Grid Operators) to interact with Advanced Distribution Management System (ADMS) and RT-RMOD. The developed algorithm was validated for multiple cases of weather events using a real-world, off-grid microgrid system modeled in a real-time simulator, sensor data, and software tools.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Physics vs structure: A systematic benchmark of learning strategies for multi-zone building thermal dynamics

Recent advances in physics-informed and data-driven machine learning promise improved thermal models for advanced building control, yet there is limited quantitative evidence on when added physics structure and architectural complexity are beneficial. Here, this work presents a systematic benchmark of five representative system identification methods for modeling multi-zone building thermal dynamics: linear state-space models, multi-layer perceptrons, neural state-space models, neural ordinary differential equations, and physically-consistent neural networks. The methods are evaluated across multiple data regimes and zone coupling strategies. Using a high-fidelity multi-zone commercial building emulator, we examine short-term and long-term prediction accuracy, computational efficiency, and ease of development. Our results reveal critical trade-offs between prediction performance, model complexity, and physical consistency. We demonstrate that decoupled, nonlinear black-box models consistently outperform coupled physics-constrained architectures in both predictive accuracy and out-of-distribution robustness in majority of the test cases for the building type considered in the study. Our findings quantify the cost of complexity in building thermal modeling and provide concrete, actionable, scenario-based guidelines for selecting model classes for control-oriented applications.

Building thermal modeling↗