Results for “simulation modeling training analysis” — Page 4

DOE OSTI2022

High Flux Isotope Reactor Low Enriched Uranium U-10Mo Fuel Design Parameters

Activities to convert the HFIR from HEU to LEU are ongoing as part of the US Department of Energy (DOE) National Nuclear Security Administration (NNSA) nuclear nonproliferation mission. Design activities to study the conversion of HFIR from HEU to LEU fuel explored different fuel design features and shapes with a uranium-molybdenum (U-10Mo) monolithic alloy fuel. This high-density alloy contains 90 wt % uranium and 10 wt % molybdenum and has a uranium density of 15.318gU/cm 3 . The goal of these studies is to generate several candidate HFIR LEU fuel designs of varying fuel fabrication complexity that meet the current HEU performance metrics and safety requirements. Recent advancements in modeling and simulation tools and design methods enabled a thorough analysis of the available design space with U-10Mo fuel. A surrogate model used this analysis as training data to quickly determine the performance of a design given specific design parameters. An optimization module used this surrogate model to quickly search this multidimensional search space given specific desired performance characteristics. This approach was made possible by the large available design space with U-10Mo fuel. Shift, a Monte Carlo tool optimized for high-performance computing (HPC) architectures, was used for faster calculation and better data management for reactor physics simulations. Once most of these design studies were complete, a new suite called the Python HFIR Analysis and Measurement Engine (PHAME) was developed to connect all fuel design analysis steps, making design studies more efficient and reproducible. The post-processing capabilities of these new tools are leveraged for the information provided herein. Leveraging these tools, several candidate fuel designs were selected with varying levels of feature complexity and reactor performance. This report provides design feature details for four selected HFIR LEU U-10Mo fuel designs and their corresponding performance and safety metrics. Nominal best-estimate design parameters and irradiation conditions, including fission rate densities, power densities, heat fluxes, and cumulative fission densities, are provided. Simulations show that the high uranium density of U-10Mo fuel provides a large potential design space that enables various LEU designs to meet HEU core performance metrics and safety requirements with a power increase from 85 MW (HEU) to 95 MW or 100 MW (LEU).

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

NASA NTRSDate not supplied

Simulation of the Aerosol Size Distribution Using a Neural Network Surrogate for the Modal Aerosol Module (MAM7)

One objective of atmospheric simulations is to quantify the distribution of aerosols and their properties. Accurate parameterizations of the processes governing aerosol mass, particle number, and particle size distribution are important for predicting the Earth’s net radiative balance and aerosol-cloud interactions. The Modal Aerosol Module (MAM7) is a two-moment aerosol model that simulates mass, number, and size distribution of seven modes comprised of internally mixed aerosol species. The two-moment scheme adds significant computational expense but allows for the prediction of varying particle size distribution relative to the bulk method which predicts only total mass. In this work, we developed a neural network surrogate model for MAM7 (MAMnet) to predict the aerosol number concentration in NASA’s Global Earth Observing System (GEOS) without adding prohibitive computational expense. MAMnet, can be driven by output from a single moment, mass-based, aerosol scheme (Goddard Chemistry Aerosol and Radiation model (GOCART)) or from reanalysis products (Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2)). MAMnet was trained using number concentrations from a 5-year GEOS/MAM7 simulation at 1-degree horizontal resolution and using the total mass calculated across modes as inputs, as well as temperature and air density. The model architecture for MAMnet was based on AlexNet, the 2012 winner of the ImageNet Large Scale Visual Recognition Challenge. While some modifications were necessary to accommodate our problem, important aspects of the network were preserved. MAMnet was able to reproduce zonal dynamics and spatial distributions of the aerosol number concentration however predictability in the upper troposphere was poor.

Katherine H Breen↗

DOE OSTI2020

DFT Accurate Interatomic Potential for Molten NaCl from Machine Learning

Molten alkali chloride salts are a critical component in concentrated solar power and nuclear applications. Despite their ubiquity, the extreme chemical reactivity of molten alkali chlorides at high temperatures has presented a significant challenge in characterizing atomic structures and dynamic properties experimentally. In this work, we we investigate molten NaCl by performing high temperature molecular dynamics simulations using a Gaussian Approximation Potential (GAP) trained on Density Functional Theory (DFT) datasets. Our GAP model, trained with a meager 1000 atomic configurations, arrives at near ab initio accuracy with a mean absolute error of 1.5 meV/atom thus enabling fast analysis of high temperature salt properties on large length (5000 ion pairs) and time (> 1ns) scales currently inaccessible to ab initio simulations. Calculated structure factors and diffusion constants from our GAP model simulations show excellent agreement with experiments. Our results indicate that GAP models are able to capture the many-body interactions required to accurately model ionic-systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

Enabling scientific machine learning in MOOSE using Libtorch

A neural-network-based machine learning interface has been developed for the Multiphysics Object-Oriented Simulation Environment (MOOSE). The interface relies on Libtorch, the C++ front-end of PyTorch, and enables an online interaction between modern machine learning algorithms and all the existing simulation, modeling, and analysis processes available in MOOSE. New capabilities in MOOSE include the native generation and training of artificial neural networks together with options to load pretrained neural networks in TorchScript format. Furthermore, the MOOSE stochastic tools module (MOOSE-STM) has been enhanced with neural network-based surrogate and reduced-order model generation options for efficient stochastic analyses. Lastly, a reinforcement learning capability has been added to MOOSE-STM for the interactive control and optimization of complex multiphysics problems.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2016

Promoting A-Priori Interoperability of HLA-Based Simulations in the Space Domain: The SISO Space Reference FOM Initiative

Distributed and Real-Time Simulation plays a key-role in the Space domain being exploited for missions and systems analysis and engineering as well as for crew training and operational support. One of the most popular standards is the 1516-2010 IEEE Standard for Modeling and Simulation (M&S) High Level Architecture (HLA). HLA supports the implementation of distributed simulations (called Federations) in which a set of simulation entities (called Federates) can interact using a Run-Time Infrastructure (RTI). In a given Federation, a Federate can publish and/or subscribes objects and interactions on the RTI only in accordance with their structures as defined in a FOM (Federation Object Model). Currently, the Space domain is characterized by a set of incompatible FOMs that, although meet the specific needs of different organizations and projects, increases the long-term cost for interoperability. In this context, the availability of a reference FOM for the Space domain will enable the development of interoperable HLA-based simulators for related joint projects and collaborations among worldwide organizations involved in the Space domain (e.g. NASA, ESA, Roscosmos, and JAXA). The paper presents a first set of results achieved by a SISO standardization effort that aims at providing a Space Reference FOM for international collaboration on Space systems simulations.

Moller, Bjorn↗

DOE OSTI2016

On Improving Analytical Models of Cosmic Reionization for Matching Numerical Simulations

The methods for studying the epoch of cosmic reionization vary from full radiative transfer simulations to purely analytical models. While numerical approaches are computationally expensive and are not suitable for generating many mock catalogs, analytical methods are based on assumptions and approximations. We explore the interconnection between both methods. First, we ask how the analytical framework of excursion set formalism can be used for statistical analysis of numerical simulations and visual representation of the morphology of ionization fronts. Second, we explore the methods of training the analytical model on a given numerical simulation. We present a new code which emerged from this study. Its main application is to match the analytical model with a numerical simulation. Then, it allows one to generate mock reionization catalogs with volumes exceeding the original simulation quickly and computationally inexpensively, meanwhile reproducing large scale statistical properties. These mock catalogs are particularly useful for CMB polarization and 21cm experiments, where large volumes are required to simulate the observed signal.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2016

ON IMPROVING ANALYTICAL MODELS OF COSMIC REIONIZATION FOR MATCHING NUMERICAL SIMULATION

The methods for studying the epoch of cosmic reionization vary from full radiative transfer simulations to purely analytical models. While numerical approaches are computationally expensive and are not suitable for generating many mock catalogs, analytical methods are based on assumptions and approximations. We explore the interconnection between both methods. First, we ask how the analytical framework of excursion set formalism can be used for statistical analysis of numerical simulations and visual representation of the morphology of ionization fronts. Second, we explore the methods of training the analytical model on a given numerical simulation. We present a new code which emerged from this study. Its main application is to match the analytical model with a numerical simulation. Then, it allows one to generate mock reionization catalogs with volumes exceeding the original simulation quickly and computationally inexpensively, meanwhile reproducing large-scale statistical properties. These mock catalogs are particularly useful for cosmic microwave background polarization and 21 cm experiments, where large volumes are required to simulate the observed signal.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRSDate not supplied

Hover and Forward Flight Performance Modeling of the Ingenuity Mars Helicopter

In 2015, NASA’s Jet Propulsion Laboratory partnered with Ames Research Center, Langley Research Center, and AeroVironment to develop Ingenuity, a small coaxial helicopter capable of flying within Mars’ unique atmospheric conditions. Ingenuity was successfully deployed from its protective shroud on the underside of the Mars 2020 Perseverance Rover and has flown 17 flights on Mars as of December 2021. A number of rotorcraft analysis tools were utilized, and a series of experimental tests were performed to ready Ingenuity for its launch with the Perseverance Rover in July 2020. In this paper, RotCFD, a Reynolds-averaged Navier-Stokes flow solver, is used to model Ingenuity in hover and forward flight for the purposes of validating tools to aid in the development of a future generation of Mars rotorcraft. The results from the RotCFD modeling are benchmarked against results from hover performance tests of the Ingenuity prototype in the 25-Foot Space Simulator at the Jet Propulsion Laboratory and are also compared to hover and forward flight predictions made by CAMRAD II, a well-known comprehensive rotorcraft analysis code. Surrogate performance models are trained to obtain a set of trimmed rotor settings for Ingenuity at different forward flight speeds, which are then used as inputs for the RotCFD forward flight simulations. Additionally, a study of the airframe-rotor interaction and a study of the aerodynamics of the individual airframe components of Ingenuity in forward flight are performed. Finally, to better understand performance predictions by RotCFD and CAMRAD II, a study is conducted on how sectional angles of attack in each code vary with radial station and azimuth.

Hover↗

DOE OSTI2024

Bayesian analysis of (3 +1)⁢D relativistic nuclear dynamics with the RHIC beam energy scan data

This work presents a Bayesian inference study for relativistic heavy-ion collisions in the beam energy scan program at the BNL Relativistic Heavy-Ion Collider. The theoretical model simulates event-by-event (3+1)-dimensional [(3+1)⁢D] collision dynamics using hydrodynamics and hadronic transport theory. We analyze the model's 20-dimensional posterior distributions obtained using three model emulators with different accuracy and demonstrate the essential role of training an accurate model emulator in the Bayesian analysis. Our analysis provides robust constraints on the quark-gluon plasma's transport properties and various aspects of (3+1)⁢D relativistic nuclear dynamics. By running full model simulations with 100 parameter sets sampled from the posterior distribution, we make predictions for p T -differential observables and estimate their systematic theory uncertainty. Here, a sensitivity analysis is performed to elucidate how individual experimental observables respond to different model parameters, providing useful physics insights into the phenomenological model for heavy-ion collisions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

NASA NTRSDate not supplied

Seamless Integration of Multibody and Orbital Dynamics to Support Human Spaceflight Simulations

At the NASA Johnson Space Center (JSC), modeling capability requirements are often driven by the need to simulate upcoming mission scenarios for analysis, procedures development, and training. In preparation for free-flyer vehicle capture, maneuvers, and release, it was determined that the multibody dynamics capabilities provided by MBDyn and the orbital dynamics capabilities supplied by JEOD needed to be merged to properly simulate the dynamic behaviors expected on-orbit. After initial attempts to combine the two packages proved cumbersome, it was decided that a more refined solution was required to facilitate the interface between these two packages. The resulting interface software package was named MBJEOD. The primary design goals for this new package were to leave MBDyn and JEOD untouched, to simplify and clarify simulation initialization for users, and to support any topology or on-orbit operational scenario that future missions might require. This paper provides a detailed overview of MBJEOD and the approach taken to seamlessly integrate the multibody dynamics of MBDyn with the orbital dynamics of JEOD. Additional attention is given to the numerical considerations required for MBJEOD and the simplifying assumptions leveraged to provide for real-time simulation needs. Verification and validation are addressed and current MBJEOD applications at the NASA JSC are discussed.

modeling↗

NASA NTRSDate not supplied

Seamless Integration of Multibody and Orbital Dynamics to Support Human Spaceflight Simulations

At the NASA Johnson Space Center (JSC), modeling capability requirements are often driven by the need to simulate upcoming mission scenarios for analysis, procedures development, and training. In preparation for free-flyer vehicle capture, maneuvers, and release, it was determined that the multibody dynamics capabilities provided by MBDyn and the orbital dynamics capabilities supplied by JEOD needed to be merged to properly simulate the dynamic behaviors expected on-orbit. After initial attempts to combine the two packages proved cumbersome, it was decided that a more refined solution was required to facilitate the interface between these two packages. The resulting interface software package was named MBJEOD. The primary design goals for this new package were to leave MBDyn and JEOD untouched, to simplify and clarify simulation initialization for users, and to support any topology or on-orbit operational scenario that future missions might require. This paper provides a detailed overview of MBJEOD and the approach taken to seamlessly integrate the multibody dynamics of MBDyn with the orbital dynamics of JEOD. Additional attention is given to the numerical considerations required for MBJEOD and the simplifying assumptions leveraged to provide for real-time simulation needs. Verification and validation are addressed and current MBJEOD applications at the NASA JSC are discussed.

modeling↗

DOE OSTI2023

Iterative sampling of expensive simulations for faster deep surrogate training

Deep neural network (DNN) surrogates of expensive physics simulations are enabling a rapid change in the way that common experimental design and analysis tasks are approached. Surrogate models allow simulations to be performed in parallel and separately from downstream tasks, thereby enabling analyses that would be impossible with the simulation in-the-loop; surrogates based on DNNs can effectively emulate diverse non-scalar data of the types collected in fusion and laboratory-astrophysics experiments. The challenge is in training the surrogate model, for which large ensembles of physics simulations must be run, preferably without wasting computational effort on uninteresting simulations. Here, in this paper, we present an iterative sampling scheme that can preferentially propose simulations in interesting regions of parameter space without neglecting unexplored regions, allowing high-quality and wide-ranging surrogate models to be trained using 2–3 times fewer simulations compare to space-filling designs. Our approach uses an explicit importance function defined on the simulation output space, balanced against a measure of simulation density which serves as a proxy for surrogate accuracy. It is easy to implement and can be tuned to find interesting simulations early in the study, allowing surrogates to be trained quickly and refined as new simulations become available; this represents an important step towards the routine generation of deep surrogate models quickly enough to be truly relevant to experimental work.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2026

Data and scripts associated with a manuscript analyzing ELM-FATES parameter sensitivity under pre-fire and postfire scenarios using machine learning

NOTE: The manuscript associated with this data package is currently in review. The data may be revised based on reviewer feedback. Upon manuscript acceptance, this data package will be updated with the final dataset and additional metadata. This data package is associated with the manuscript “Fire Severity-Dependent Shifts in Vegetation Parameter Sensitivity: A Pre- and Post-Fire Analysis Using ELM-FATES and Explainable AI” submitted to Journal of Advances in Modeling Earth Systems (Zahura et al. 2026). The study examines vegetation physiological parameters controlling pre-fire and post-fire vegetation dynamics. To support this analysis, 73 vegetation parameters in Functionally Assembled Terrestrial Ecosystem Simulator (FATES) (Fisher et al., 2018) , which is coupled with E3SM (Energy Exascale Earth System Model) land model (ELM, ELM-FATES), were perturbed using a Sobol sequence to generate 1,024 ensemble members for two plant functional types: needleleaf evergreen extratropical trees (NEET) and C3 grass. Simulations were conducted for the pre-fire period (2016) and post-fire period (2018–2023). Burn severity was represented by modifying the Nesterov index in FATES to 75,000, 150,000, and 300,000 for low, moderate, and high severity, respectively. A no-fire scenario was also included. Simulations were performed for 16 grid cells in the American River Watershed across different burn severities and plant functional types. XGBoost (eXtreme Gradient Boosting) models were trained using the parameter ensembles and ELM-FATES-simulated outputs, including leaf area index (LAI), gross primary productivity (GPP), aboveground biomass, vegetation evaporation, transpiration, and soil evaporation. Models were trained separately for each year and burn severity, followed by SHAP (SHapley Additive exPlanations) analysis to identify changes in dominant parameters after fire disturbance. For details on how to navigate data packages generated by this project, see https://data.ess-dive.lbl.gov/portals/PNNLRiverCorridorSFA/About. The data package contains the ELM-FATES simulation data. The scripts and data related to the analysis will be added later. The inputs and outputs from ELM-FATES are inside the “FATES” folder. “FATES_domain_surface” contains the domain and surface netcdfs that were used to run ELM-FATES in the study area. “FATES_parameters” contains the 1024 ensembles that were generated using Sobol sequence. “FATES_outputs” folder contains ELM-FATES simulated variables. All files are .csv and .nc (NetCDF).

Aboveground biomass↗

DOE OSTI2016

Co-Optimization of CO 2 -EOR and Storage Processes in Mature Oil Reservoirs

This article presents an optimization methodology for CO 2 enhanced oil recovery in partially depleted reservoirs. A field-scale compositional reservoir flow model was developed for assessing the performance history of an active CO 2 flood and for optimizing both oil production and CO 2 storage in the Farnsworth Unit (FWU), Ochiltree County, Texas. A geological framework model constructed from geophysical, geological, and engineering data acquired from the FWU was the basis for all reservoir simulations and the optimization method. An equation of state was calibrated with laboratory fluid analyses and subsequently used to predict the thermodynamic minimum miscible pressure (MMP). Initial history calibrations of primary, secondary and tertiary recovery were conducted as the basis for the study. After a good match was achieved, an optimization approach consisting of a proxy or surrogate model was constructed with a polynomial response surface method (PRSM). The PRSM utilized an objective function that maximized both oil recovery and CO 2 storage. Experimental design was used to link uncertain parameters to the objective function. Control variables considered in this study included: water alternating gas cycle and ratio, production rates and bottom-hole pressure of injectors and producers. Other key parameters considered in the modeling process were CO 2 purchase, gas recycle and addition of infill wells and/or patterns. The PRSM proxy model was ‘trained’ or calibrated with a series of training simulations. This involved an iterative process until the surrogate model reached a specific validation criterion. A sensitivity analysis was first conducted to ascertain which of these control variables to retain in the surrogate model. A genetic algorithm with a mixed-integer capability optimization approach was employed to determine the optimum developmental strategy to maximize both oil recovery and CO 2 storage. The proxy model reduced the computational cost significantly. The validation criteria of the reduced order model ensured accuracy in the dynamic modeling results. The prediction outcome suggested robustness and reliability of the genetic algorithm for optimizing both oil recovery and CO 2 storage. The reservoir modeling approach used in this study illustrates an improved approach to optimizing oil production and CO 2 storage within partially depleted oil reservoirs such as FWU. Lastly, this study may serve as a benchmark for potential CO 2 –EOR projects in the Anadarko basin and/or geologically similar basins throughout the world.

02 PETROLEUM↗

DOE OSTI2026

Data, model inputs, and analysis scripts associated with a manuscript on stream intermittency controls across spatial scales in Pacific Northwest watersheds

NOTE: The manuscript associated with this data package is currently in review. The data may be revised based on reviewer feedback. Upon manuscript acceptance, this data package will be updated with the final dataset and additional metadata. This data package is associated with the manuscript "Hydroclimatic Memory and Watershed Template Shape Stream Intermittency: Multi-scale Attribution Using Process-based Simulation and Explainable ML" by Niroula et al. (2026), submitted to Water Resources Research (WRR). The study investigates the dominant controls on stream intermittency across local, reach, and watershed scales using a coupled process-based simulation and explainable machine-learning framework. Long-term daily simulations from the Advanced Terrestrial Simulator (ATS) were used to generate wetness states and ponded-depth responses over river-corridor cells. These ATS outputs were then aggregated across scales and used to train XGBoost (eXtreme Gradient Boosting) models. SHAP (SHapley Additive exPlanations) was applied to quantify the relative importance of hydroclimatic forcings, watershed template attributes, and antecedent-memory effects in shaping intermittency behavior. The analysis is carried out for three contrasting Pacific Northwest watersheds: Oak Creek (OCW), American River Watershed (ARW), and H.J. Andrews (HJA). Across these testbeds, the package contains ATS-ready watershed inputs, ATS run configuration and selected output files, model-evaluation data products, intermittency-analysis datasets, machine-learning target-feature tables, SHAP outputs, and notebooks used to organize, analyze, and visualize results. At a high level, the package documents a workflow in which ATS provides the physically based simulation backbone and explainable machine learning is used as a post-processing attribution tool. The contents are intended to support interpretation of the manuscript figures and results, provide context for how intermittency metrics were generated at multiple scales, and preserve the key artifacts needed to understand and reuse the analysis workflow. The package contains a high-level directory summary file (`summary.txt`) and four main content folders (1) `evaluation_plots` contains evaluation figures and supporting evaluation datasets; (2) `intermittency_plots` contains intermittency-focused analysis notebook and prepared datasets; (3) `ml-training-and-shap_values_plots` contains ML training inputs, SHAP outputs, and figure-generation notebooks; and (4) `watershed_mesh_and_ats_input` contains ATS model setup materials, forcing inputs, geometry, and selected run files. More specifically, the `evaluation_plots` folder contains the notebook used for ATS evaluation plotting and site-specific evaluation datasets. These include evapotranspiration and water-balance products for three watersheds, as well as an Oak Creek field-measurement discharge file. The `intermittency_plots` folder contains the notebook used for intermittency analysis and the prepared datasets used to analyze intermittent and non-intermittent wetness behavior across the study watersheds. The `ml-training-and-shap_values_plots` folder contains notebooks and outputs for the machine-learning and explainability workflow. This includes the main XGBoost and SHAP notebook(s), a beeswarm plotting notebook, target-feature tables for machine-learning training, SHAP summary tables, and per-sample SHAP value archives. The `watershed_mesh_and_ats_input` folder contains ATS-related watershed inputs and supporting materials. This includes mesh and shape products, ATS-readable LAI and meteorological forcing inputs, selected ATS spinup and transient-run files, and a watershed workflow example notebook. Subdirectories are organized by watershed where applicable.All files are .cpg (codepage files), .csv (comma-separated values), .dbf (database files), .exo (Exodus mesh format), .h5 (HDF5 format), .ipynb (Jupyter notebooks), .pkl (Python pickle), .prj (projection files), .sh (shell scripts), .shp (shapefile geometry), .shx (shapefile index), .txt (text files), or .xml (markup data).

Advanced Terrestrial Simulator↗

DOE OSTI2018

Gradient-based optimization for regression in the functional tensor-train format

Predictive analysis of complex computational models, such as uncertainty quantification (UQ), must often rely on using an existing database of simulation runs. In this paper we consider the task of performing low-multilinear-rank regression on such a database. Specifically we develop and analyze an efficient gradient computation that enables gradient-based optimization procedures, including stochastic gradient descent and quasi-Newton methods, for learning the parameters of a functional tensor-train (FT). We compare our algorithms with 22 other nonparametric and parametric regression methods on 10 real-world data sets and show that for many physical systems, exploiting low-rank structure facilitates efficient construction of surrogate models. Here, we use a number of synthetic functions to build insight into behavior of our algorithms, including the rank adaptation and group-sparsity regularization procedures that we developed to reduce overfitting. Finally we conclude the paper by building a surrogate of a physical model of a propulsion plant on a naval vessel.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2017

Electric train energy consumption modeling

For this paper we develop an electric train energy consumption modeling framework considering instantaneous regenerative braking efficiency in support of a rail simulation system. The model is calibrated with data from Portland, Oregon using an unconstrained non-linear optimization procedure, and validated using data from Chicago, Illinois by comparing model predictions against the National Transit Database (NTD) estimates. The results demonstrate that regenerative braking efficiency varies as an exponential function of the deceleration level, rather than an average constant as assumed in previous studies. The model predictions are demonstrated to be consistent with the NTD estimates, producing a predicted error of 1.87% and -2.31%. The paper demonstrates that energy recovery reduces the overall power consumption by 20% for the tested Chicago route. Furthermore, the paper demonstrates that the proposed modeling approach is able to capture energy consumption differences associated with train, route and operational parameters, and thus is applicable for project-level analysis. The model can be easily implemented in traffic simulation software, used in smartphone applications and eco-transit programs given its fast execution time and easy integration in complex frameworks.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2026

A hybrid Penman-Monteith and machine learning model for simulating evapotranspiration and its components

Integrating physical processes with machine learning has advanced evapotranspiration (ET) simulation, yet most hybrid models fail to partition total ET into its components: soil evaporation (E) and vegetation transpiration (T). This study introduces Residual Neural Network–Penman–Monteith (RNN-PM), a novel hybrid dual-source ET model designed to overcome this limitation. The model synergizes the physically-based Penman–Monteith framework with three specialized residual neural networks trained to estimate key conductance parameters (canopy conductance, soil surface conductance, and aerodynamic conductance). Furthermore this explicit parameterization allows for the direct partitioning of total ET. Validation at National Ecological Observatory Network (NEON) flux sites using high-frequency partitioned E and T shows that RNN-PM reliably reproduces ET and the transpiration fraction (T/ET). For ET, the model achieves an average Kling–Gupta efficiency (KGE) of 0.89 and a root-mean-square error (RMSE) of 0.55 mm/day; for T/ET, the KGE is 0.87 with an RMSE of 0.06. Furthermore, RNN-PM demonstrates robust generalization, accurately simulating ET and its components well beyond the initial training dataset, even under extreme climatic conditions. This study extended the analysis by comparing the RNN-PM model with seven established dual-source ET models. The results indicate that RNN-PM outperforms both conventional machine learning models and purely physical process-based models in simulating ET components in most cases. Among the purely physical process-based dual-source models, those based on surface temperature decomposition showed improved performance as the leaf area index (LAI) decreased when evaluated against high-frequency ET component datasets. In contrast, the performance of conductance-based dual-source models declined with decreasing LAI. Although purely machine learning-based models can produce relatively accurate simulations of ET components, they often exhibit limited generalization capability, an issue that the RNN-PM model effectively overcomes. Ultimately, the RNN-PM model represents a significant advance in simulating ET components, offering a novel and scalable approach for improving the representation of land–atmosphere interactions in Earth system models.

54 ENVIRONMENTAL SCIENCES↗