Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Hybrid model”

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 685 records · Page 38
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Data-driven recovery of hidden physics in reduced order modeling of fluid flows

In this article, we introduce a modular hybrid analysis and modeling (HAM) approach to account for hidden physics in reduced order modeling (ROM) of parameterized systems relevant to fluid dynamics. The hybrid ROM framework is based on using first principles to model the known physics in conjunction with utilizing the data-driven machine learning tools to model the remaining residual that is hidden in data. This framework employs proper orthogonal decomposition as a compression tool to construct orthonormal bases and Galerkin projection (GP) as a model to build the dynamical core of the system. Our proposed methodology hence compensates structural or epistemic uncertainties in models and utilizes the observed data snapshots to compute true modal coefficients spanned by these bases. The GP model is then corrected at every time step with a data-driven rectification using a long short-term memory (LSTM) neural network architecture to incorporate hidden physics. A Grassmannian manifold approach is also adopted for interpolating basis functions to unseen parametric conditions. The control parameter governing the system's behavior is thus implicitly considered through true modal coefficients as input features to the LSTM network. The effectiveness of the HAM approach is then discussed through illustrative examples that are generated synthetically to take hidden physics into account. Furthermore, our approach thus provides insights addressing a fundamental limitation of the physics-based models when the governing equations are incomplete to represent underlying physical processes.

42 ENGINEERING↗
DOE OSTI2022

Hybrid Solid Oxide Fuel Cell/Gas Turbine Model Development for Electric Aviation

A thermodynamic model was developed and validated to analyze a high-performance solid oxide fuel cell and gas turbine (SOFC-GT) hybrid power system for electric aviation. This study used a process simulation software package (ProMax) to study the role of SOFC design and operation on the feasibility and performance of the hybrid system. Standard modules, including compressor, turbine, heat exchanger, reforming reactor, and combustor were used from the ProMax tool suite while a custom module was created to simulate the SOFC stack. The model used an SOFC test data set as an input. Additional SOFC stack performance effects, such as pressure, temperature, and utilization of air and fuel, were added from open source data. System performance predictors were SOFC specific power, fuel-to-electricity conversion efficiency, and hybrid system efficiency. Using these input data and predictors, a static thermodynamic performance model was created that can be modified for different system configurations and operating conditions. Prior to creating the final aircraft performance model, initial demonstration models were developed to validate output results. We used the NASA SOFC model as a benchmark, which was created with their Numerical Propulsion System Simulator (NPSS) software framework. Our output results matched within 1% of both the NASA model and open source SOFC performance data. With confidence gained in the accuracy of this model, a 1-MW SOFC-GT hybrid power system was constructed for an aircraft propulsion concept. Overall hybrid system efficiencies of > 75% FTE were observed during standard 36,000 feet cruise flight conditions.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2024

Multi-Timescale Optimal Operation Framework for Integrated Economic and Reliability Analysis of Hybrid Power Plants

This paper introduces a hierarchical modeling framework for hybrid power plants (HPP) to facilitate the operation of HPP in power systems similar to conventional generators (Congens) in the integrated multi-timescale optimal operation framework. To consider the uncertainties of HPP renewable power in the day-ahead scheduling, distributionally robust optimization (DRO) is used. To ensure that the state-of-charge (SOC) of energy storage systems in HPPs aligns closely with the planned value for long-term reliability, real-time SOC management is incorporated. In addition, an adjustable real-time control is designed for the robust delivery of HPP real-time services. Case studies performed on a revised IEEE 39-bus system demonstrate the effectiveness of the proposed framework for HPP operation. Simulation results highlight that the proposed framework not only can help operators schedule HPP similar to Congens in varying weather conditions but can also maintain the frequency reliability of the system.

frequency stability↗
NASA NTRS1974

Small scale model static acoustic investigation of hybrid high lift systems combining upper surface blowing with the internally blown flap

A static acoustic and propulsion test of a small radius Jacobs-Hurkamp and a large radius Flex Flap combined with four upper surface blowing (USB) nozzles was performed. Nozzle force and flow data, flap trailing edge total pressure survey data, and acoustic data were obtained. Jacobs-Hurkamp flap surface pressure data, flow visualization photographs, and spoiler acoustic data from the limited mid-year tests are reported. A pressure ratio range of 1.2 to 1.5 was investigated for the USB nozzles and for the auxiliary blowing slots. The acoustic data were scaled to a four-engine STOL airplane of roughly 110,000 kilograms or 50,000 pounds gross weight, corresponding to a model scale of approximately 0.2 for the nozzles without deflector. The model nozzle scale is actually reduced to about .17 with deflector although all results in this report assume 0.2 scale factor. Trailing edge pressure surveys indicated that poor flow attachment was obtained even at large flow impingement angles unless a nozzle deflector plate was used. Good attachment was obtained with the aspect ratio four nozzle with deflector, confirming the small scale wind tunnel tests.

Cole, T. W.↗
DOE OSTI2025

Control of core–shell nanoparticles properties through plasma synthesis: a computational study

The improved properties of core–shell nanoparticles (CSNPs) over homogeneous nanoparticles (NPs) have expanded and diversified the applications of these nanomaterials. However, controlling the properties of CSNPs can be a challenging task. Low temperature plasmas have proven to be an effective method of producing NPs with uniform size and morphology, and high yield. That said, NP transport and growth dynamics are sensitive to LTP properties. We report on a computational investigation of the evolution of Ge–Si CSNP properties as a function of operating conditions through the modeling of a flowing, two-zone inductively coupled plasma (ICP) reactor. Ar/GeH 4 and Ar/SiH 4 gas mixtures were supplied to separate plasma zones at a pressure of 1 Torr to promote growth of Ge cores and Si shells. The negatively charged CSNPs are trapped electrostatically in the vicinity of the antennas where the plasma is generated and where the majority of particle growth occurs. Particles that grow to a critical size are then de-trapped by fluid drag due to neutral gas flow. A two-dimensional hybrid plasma model coupled with a three-dimensional kinetic NP transport model were utilized to resolve plasma chemistry and NP growth processes that take place on distinct timescales. The trends in CSNP properties and trapping mechanisms associated with flow rate, applied ICP power and inlet precursor fraction are discussed. While the spatial distribution of plasma produced radical species can have significant impact on the NP growth process, the NP transport dynamics are what ultimately dictates the growth environment that is unique to each particle and so determines their final dimension and composition. The key to optimizing reactor conditions involves controlling the spatial density of growth species and plasma profile as a means to tailor particle trapping dynamics suitable to produce CSNPs for a specific application.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Physics-based hybrid machine learning for critical heat flux prediction with uncertainty quantification

Critical heat flux (CHF) is a key quantity in nuclear system modeling due to its impact on heat transfer, safety margins, and reactor performance. This study develops and validates an uncertainty-aware hybrid modeling approach that combines machine learning with physics-based models to predict CHF in cases of dryout. The Biasi and Bowring empirical correlations were paired with three ML uncertainty quantification (UQ) techniques: deep neural network (DNN) ensembles, Bayesian neural networks (BNNs), and deep Gaussian processes (DGPs). A pure ML model without a base model was evaluated for comparison. Model performance was assessed under plentiful (7,350 points) and limited (9 points) training data scenarios using parity, uncertainty distributions, and calibration curves. Results show that the Biasi hybrid DNN ensemble achieved the best overall performance, with a mean absolute relative error of 1.846%, and well-calibrated uncertainty estimates. The BNN-based hybrids showed slightly higher error (2.14%) but superior uncertainty calibration. DGP models underperformed, with over 6% error and poor uncertainty calibration. All hybrid models outperformed pure machine learning configurations, demonstrating resistance against data scarcity. These findings indicate that hybrid modeling significantly improves predictive accuracy, interpretability, and resilience to data scarcity. The integration of uncertainty awareness provides actionable confidence in CHF predictions, which is vital for safety-critical decisions in nuclear applications. This hybrid approach offers a viable pathway for deploying ML models in reactor analysis tools while preserving domain knowledge and physical consistency.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRS1984

The origin of low-velocity absorption components in the Mg II resonance lines of hybrid-chromosphere stars

It is argued that the low-velocity absorption features seen in the Mg II resonance lines of seven confirmed and three probable hybrid-chromosphere stars are interstellar rather than circumstellar in origin. From a comparison of radial velocities based on all available spectra in the International Ultraviolet Explorer (IUE) archives with estimates of the interstellar velocity along each line of sight, a good correlation between the observed position of the low-velocity component and the predicted interstellar feature is found. It is also shown that previous arguments in favor of the circumstellar origin of the low-velocity absorption features are either incorrect or implausible. The conclusion of the present study may modify previously proposed models of hybrid star winds which have assumed a priori that both Mg II absorption components are circumstellar.

Drake, S. A.↗
NASA NTRS2016

Challenges and Progress in Aerodynamic Design of Hybrid Wingbody Aircraft with Embedded Engines

We summarize the contributions to high-fidelity capabilities for analysis and design of hybrid wingbody (HWB) configurations considered by NASA. Specifically, we focus on the embedded propulsion concepts of the N2-B and N3-X configurations, some of the future concepts seriously investigated by the NASA Fixed Wing Project. The objective is to develop the capability to compute the integrated propulsion and airframe system realistically in geometry and accurately in flow physics. In particular, the propulsion system (including the entire engine core-compressor, combustor, and turbine stages) is vastly more difficult and costly to simulate with the same level of fidelity as the external aerodynamics. Hence, we develop an accurate modeling approach that retains important physical parameters relevant to aerodynamic and propulsion analyses for evaluating the HWB concepts. Having the analytical capabilities at our disposal, concerns and issues that were considered to be critical for the HWB concepts can now be assessed reliably and systematically; assumptions invoked by previous studies were found to have serious consequences in our study. During this task, we establish firmly that aerodynamic analysis of a HWB concept without including installation of the propulsion system is far from realistic and can be misleading. Challenges in delivering the often-cited advantages that belong to the HWB are the focus of our study and are emphasized in this report. We have attempted to address these challenges and have had successes, which are summarized here. Some can have broad implications, such as the concept of flow conditioning for reducing flow distortion and the modeling of fan stages. The design optimization capability developed for improving the aerodynamic characteristics of the baseline HWB configurations is general and can be employed for other applications. Further improvement of the N3-X configuration can be expected by expanding the design space. Finally, the support of the System Analysis and Integration Element under the NASA Fixed Wing Project has enabled the development and helped deployment of the capabilities shown in this report.

design optimization↗
DOE OSTI2022

Artificial Intelligence/Machine Learning Technologies for Advanced Reactors (Workshop Summary Report)

A workshop on artificial intelligence and machine learning (AI/ML) for advanced reactors (AR) was held October 5-6, 2021. The workshop was to be attended in-person at ANL but COVID restrictions forced the workshop to go virtual. The objectives of the workshop were to identify the most promising AI/ML opportunities for improving advanced reactor design, optimizing plant performance, and enhancing economic competitiveness and to develop an understanding of the scientific, engineering and licensing challenges facing their application. The workshop planning committee included GAIN, EPRI and NEI and members of three national laboratories (ANL, INL, and ORNL). The workshop was attended by more than 200 individuals representing academic and scientific institutions and the nuclear power industry. The definition put forth for an AI/ML system was one that perceives its environment and takes actions that maximize its chance of achieving its goals. In this report AI/ML refers to next generation algorithms that include deep learning, statistical analysis and data analytics and associated scientific computing and their potential application to the design, licensing, operation and maintenance of ARs. These methods typically incorporate models built from process data and may also include data generated by simulations that represent the behavior of a system. The workshop was organized in response to the growing interest in application of AI/ML for improving the economic competitiveness of nuclear energy. Increasingly more resources are being allocated to investigating the benefits of AI/ML methods. The DOE created the Artificial Intelligence & Technology Office to promote their development. And within the Office of Nuclear Energy, resources have been allocated to explore and understand the potential benefits of AI/ML. Additionally, the national laboratories are strategically positioned with DOE computing facilities such as Summit, Perlmutter, Aurora and Frontier that support large-scale simulations, hybrid HPC models with AI surrogates, and the exploration of new types of generative models emerging from multi-model data streams and sources. The workshop was organized with members of the AR community to understand the effort and to identify the level of interest and progress in this emerging technology. The workshop discussions focused on identifying opportunities for AI/ML across diverse areas of the nuclear industry and identifying current scientific and engineering challenges for advanced reactors that might be addressed through transformational uses of AI/ML. Discussion panels focused on four high-interest technical domains for advanced reactors: design, maintenance and operations, energy storage, and materials. The results of those discussions are summarized in this report. This includes opportunities that were identified for exploiting AI techniques and methods to improve the efficacy and efficiency of reactor analysis and to improve the operation and optimization of advanced reactors. Advanced reactor developers expressed an interest in learning more about AI/ML methods and their application. This included understanding whether ML methods can provide an advantage over existing nonlinear data regression methods for collapsing high-fidelity simulation results into faster running models. A consensus emerged that AR advances planned for the next decade will benefit from the use of AI/ML tools. The need exists to understand and model complex systems across length scales and modalities. AI/ML is a tool for discovery that can yield a set of engineering principles for use by nuclear engineers, licensing bodies, and operators to solve problems in plant design, safety analyses, autonomous operation, and predictive maintenance. While AI/ML represents a new set of tools, an awareness by the nuclear community of the full potential is still in the early stages so there is a need to increase awareness. It appears that the wide-spread adoption of AI/ML tools for ARs would be facilitated by future educational workshops that describe foundational methods and capabilities and describe successful applications.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2025

Capability in Theory, Modeling, and Validation for a Range of Innovative Fusion Concepts using High-Fidelity Moment-Kinetic Models

A computational modeling capability is created and available to the fusion community to understand and design lower-cost and innovative fusion concepts. The approach uses high- fidelity kinetic, moment-kinetic, and moment models and includes sophisticated plasma- boundary interactions. A majority of fusion-relevant simulations are performed with magnetohydrodynamic models and hybrid particle-in-cell codes, with limited-fidelity electron and kinetic physics. However, in fusion configurations like Z-pinches, field-reversed- configurations, plasma jet magneto-inertial fusion, spinning mirrors, and others, kinetic effects (both electron and ions) are critical to understand the physics and design scaling into the highly kinetic regime of a burning fusion plasma. Furthermore, as present fusion machines move towards a burning plasma regime, liquid-metal blankets are needed to handle first-wall heat- flux, reduce erosion, and eventually for energy conversion and fuel breeding. The work performed under this ARPA-E BETHE Capability Team advances the state-of-the-art in modeling and understanding plasma dynamics in fusion devices and its coupling with liquid-metal dynamics. These are critical areas of research for fusion energy to become realizable. To address these complex problems, we have leveraged and extended computational capabilities through the code, Gkeyll (developed jointly with Princeton Plasma Physics Laboratory and academic partners), for kinetic and moment modeling of fusion plasmas. The Concept Teams supported by this Capability Team include the Wisconsin High-field Axisymmetric Mirror (WHAM), Centrifugal Mirror Experiment (CFME), Plasma-Jet Magneto- Inertial Fusion (PJMIF), and solid and liquid wall plasma-material interaction studies relevant to a number of fusion concepts including Zap Energy’s Z-pinch. This software is open-source and available to the fusion community as a high-fidelity tool for the design of lower-cost fusion experiments. 3D gyrokinetic simulations of WHAM are now possible for long enough time scales to understand the evolution of interchange instabilities. 3D multi-fluid simulations of CMFE at higher Mach numbers are now possible for detailed design iterations with the goal of stability. The state-of-the-art in understanding shock formation and shock mitigation regimes in merging liners for PJMIF have been furthered by our kinetic simulations. Our novel models and frameworks studying plasma-material interaction by incorporating wall emission for various solid wall materials of relevance to pulsed and steady fusion concepts have advanced the state-of-the-art in our understanding of particle fluxes, heat fluxes, and other quantities at cathodes and anodes. The results from this work may explain discrepancies between experimental and theoretical predictions of achieved current densities in pulsed concepts such as Z-pinches. Another significant contribution of this Capability Team is the development and deployment of a novel experimental platform, LEX (Liquid Electrode eXperiment), at Virginia Tech to understand liquid metal free-surface response to electromagnetic pulses. The novel experiments along with model validation quantified the effect of different materials and sizes of liquid metal droplets on the radiative power balance of fusion plasmas for pulsed concepts. Furthermore, these experiments provided mitigation strategies for violent liquid metal response for high current pulses as would be expected in fusion regimes.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

ExaWind: A multifidelity modeling and simulation environment for wind energy

We introduce the open-source ExaWind modeling and simulation environment for wind energy. The primary physics codes of ExaWind are Nalu-Wind and OpenFAST. Nalu-Wind is a wind-focused computational fluid dynamics (CFD) code that is coupled to the whole-turbine simulation code OpenFAST. The ExaWind environment was created under U.S. Department of Energy funding to achieve the highest-fidelity simulations of wind turbines and wind farms to date, with the goal of enabling disruptive changes to turbine and plant design and operation. Innovation will be gleaned through better understanding of the complex flow dynamics in wind farms, including wake evolution and the impact of wakes on downstream turbines and turbulent flow from complex terrain. High-fidelity predictive simulations employ hybrid turbulence models, geometry/boundary-layer-resolving CFD meshes, atmospheric turbulence, nonlinear structural dynamics, and fluid-structure interaction. While there is an emphasis on very high-fidelity simulations (e.g., blade resolved with full fluid-structure coupling), the ExaWind environment supports lower-fidelity modeling capabilities including actuator-line and -disk methods. Important in the development of ExaWind codes is that the codes scale well on today's largest petascale supercomputers and on the next-generation platforms that will enable exascale computing.

17 WIND ENERGY↗
NASA NTRS2016

Turbine Powered Simulator Calibration and Testing for Hybrid Wing Body Powered Airframe Integration

Propulsion airframe integration testing on a 5.75% scale hybrid wing body model us- ing turbine powered simulators was completed at the National Full-Scale Aerodynamics Complex 40- by 80-foot test section. Four rear control surface con gurations including a no control surface de ection con guration were tested with the turbine powered simulator units to investigate how the jet exhaust in uenced the control surface performance as re- lated to the resultant forces and moments on the model. Compared to ow-through nacelle testing on the same hybrid wing body model, the control surface e ectiveness was found to increase with the turbine powered simulator units operating. This was true for pitching moment, lift, and drag although pitching moment was the parameter of greatest interest for this project. With the turbine powered simulator units operating, the model pitching moment was seen to increase when compared to the ow-through nacelle con guration indicating that the center elevon and vertical tail control authority increased with the jet exhaust from the turbine powered simulator units.

Shea, Patrick R.↗
DOE OSTI2025

Combined Influence of Rotation and Scrape-Off Layer Drifts on Recycling Asymmetries in Tokamak Plasmas

Coupled 2D fluid-kinetic simulations of a DIII-D high confinement tokamak plasma show that plasma rotation coupled with drift effects near the plasma edge play a significant role in the creation of the observed poloidal distribution of neutrals. It is observed that including either drift or rotation effects enhances particle flux at the inner target in the case of ion 𝐵×∇𝐵 drift toward the 𝑋-point. However, the particle flux asymmetry is significantly higher with the combination of drifts and rotation than either effect alone. The heightened particle flux asymmetry allows for improved simulation of the strong in-out asymmetry of the Lyman-𝛼 brightness profiles measured in the experiment. Enhancement of radial transport of parallel momentum changes the upstream scrape-off layer flow pattern, increasing the fraction of deuterium flux that reaches the inboard divertor entrance while lowering that which arrives at the outboard. In conclusion, this Letter indicates that by combining drifts, rotation, and viscous coupling, existing boundary plasma models can achieve a satisfactory agreement with experimentally measured neutral asymmetries.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Process Design and Techno-Economic Analysis of the Modular Staged Pressurized Oxy-Combustion (SPOC) Power Plant for Biomass

This work describes the process design and techno-economic analysis (TEA) of the modular SPOC power plant for biomass firing and coal-biomass co-firing. Two Rankine cycles were considered: a supercritical steam cycle (242 bar, 593°C, 593°C) with 550 MWe net output and a subcritical cycle (166 bar, 566°C, 566°C) with 200 MWe net output. For both cases, 95% carbon capture was modeled, and hybrid poplar biomass was chosen to generate carbon-negative power. In addition, the supercritical 500 MWe case included a 25% biomass co-firing (carbon neutral) case. For both cycles, a 100% Powder River Basin coal firing case was used for comparison purposes. In the SPOC process, oxygen is produced via a cryogenic air separation unit (ASU) and the heat generated from the compression of air is integrated into the steam cycle and utilized for boiler feed water pre-heating. Unique to the SPOC process, the boilers are pressurized and arranged in a series-parallel configuration, with minimized flue gas recirculation. The flue gas is cooled and scrubbed in the direct-contact cooler (DCC) column, and the moisture in the flue gas is condensed, leaving the bottom of the DCC at a sufficiently high temperature such that it can be used for boiler feed water pre-heating, improving plant thermal efficiency. Following drying and purification, CO2 in the flue gas is at the purity required for storage or utilization. The performance data were obtained from process modelling via Aspen Plus®. The stream data from Aspen Plus® were used as an input for the AACE Class 5 cost study. Ultimately, the capital costs, Levelized Cost of Electricity (LCOE), and cost of CO2 captured and avoided were obtained. The HHV efficiency of the carbon negative 550 MWe supercritical SPOC case (34.8%) was clearly above those reported by NETL for the BECCS baseline cases of supercritical pulverized coal with capture (B12B, 31.5%) and the 49% biomass co-firing case with capture (PA3, 29.2%). The HHV efficiency of the carbon-negative subcritical plant is also higher than the subcritical baseline PC plant with capture (case B11B.95) presented by NETL (32% vs 29.7%). The LCOE for the SPOC 100% biomass case was similar to the LCOE for the BECCS 49% biomass with carbon capture case ($147/MWh), and the SPOC carbon neutral case LCOE was lower ($110/MWh) than the cost for the NETL baseline SC coal firing case with 90% carbon capture ($114/MWh).

Magalhaes, Duarte↗
DOE OSTI2022

Evaluating performance of different generative adversarial networks for large-scale building power demand prediction

We report as an unsupervised-learning data-driven model, Generative Adversarial Networks (GANs) have recently attracted a lot of attention for various applications. There is potential to apply GANs for large-scale building power demand prediction, which is needed for power grid operation. However, there are many GAN variations and it is unclear which GAN is suitable for this application. To answer this question, this paper identifies five promising GANs (Original GAN, cGAN, SGAN, InfoGAN, and ACGAN) and evaluates their performance for predicting building power demand at a large scale. Physics-based building energy models are developed to generate training and reference data. A new evaluation indicator that combines accuracy and reproducibility is proposed to evaluate the performance of different GANs in predicting building power demand. The results show that SGAN and InfoGAN are not suitable because they cannot control the number of generated building samples for different building types. The prediction performance among the Original GAN, cGAN, and ACGAN can vary depending on training sample sizes and number of building types. If the training sample size is sufficiently large, Original GAN and cGAN can predict building power demand more accurately than ACGAN with the same number of samples. If training samples are limited, Original GAN provides better accuracy than cGAN and ACGAN. When the number of building types increase, the prediction accuracy increases for cGAN, decreases for ACGAN, and remains the same for Original GAN. As a result, cGAN and Original GAN are recommended for large-scale building power demand prediction.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Knowledge-guided graph machine learning for spatially distributed prediction of daily discharge and nitrogen export dynamics

Spatially distributed prediction of streamflow and nitrogen export dynamics is essential for precision management of agricultural watersheds. While temporal deep learning models such as Long Short-Term Memory (LSTM) have shown strong performance at basin scales, their ability to generalize spatially is limited by insufficient representation of spatial dependencies and flow paths, particularly under data-scarce conditions. To address this gap, we propose HydroGraphNet, a knowledge-guided graph machine learning framework that integrates process-based knowledge and explicit spatial learning into temporal modeling. This framework incorporates directed graph topology to encode watershed connectivity and upstream inflows, with mass balance constraints to improve physical consistency. To enhance generalization in sparsely monitored regions, HydroGraphNet is pretrained on synthetic data generated by the SWAT+ (Soil and Water Assessment Tool Plus) model. We evaluated HydroGraphNet in the Upper Sangamon River Basin (44 HUC-12 subwatersheds, 2001–2020) against two LSTM baselines: a lumped basin-level model and a distributed variant. When benchmarked on SWAT+ simulations in pretraining, HydroGraphNet improved test NSEs by 8.9% (discharge) and 13.7% (NO₃–N load) in temporal extrapolation, and by 27.1% and 34.7% in spatial extrapolation, relative to the Lumped LSTM baseline. After fine-tuning with USGS monitoring data, the model achieved mean test NSE (KGE) scores of 0.768 (0.861) for discharge and 0.626 (0.664) for NO₃–N load, substantially outperforming baselines. Attribution analysis further highlighted the importance of upstream inflow representation and graph-based spatial learning in capturing cross-subwatershed dependencies. The model also reproduced seasonal hydrological and biogeochemical patterns consistent with known processes, demonstrating its robustness and process fidelity for spatially distributed prediction. Altogether, HydroGraphNet advances the integration of physical knowledge and spatially explicit learning in hydrological modeling, offering a generalizable framework for distributed modeling to support spatially targeted water quality management in data-scarce watersheds.

54 ENVIRONMENTAL SCIENCES↗