Search NASA⌕ Search

SEARCH · Search NASA

Results for “underperformances”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Field Experience Detecting PV Underperformance in Real Time Using Existing Instrumentation

Maintenance at large-scale photovoltaic plants employs a mix of preventative and corrective maintenance practices. Large outages, such as an inverter tripping offline, are often easy to detect. More subtle sub-inverter faults and failures can accumulate and go unnoticed for months or years. A software-based fault detection method has been developed to analyze commonly measured data from large-scale PV plants for more timely detection of subtle underperformance. The method has been demonstrated on eight datasets from large-scale plants with high accuracy of detection. Results are validated using aerial infrared scanning. String outages are detected with a true positive rate of 73 percent and tracker issues are detected with a true positive rate of 88 percent. The developed method can be uniformly applied to photovoltaic plants across a range of scales and configurations to assess performance, quickly detect underperformance, and determine the source and location of failures. The results inform and improve operations and maintenance at PV plants, ultimately aiding in improved affordability, reliability, availability, and resiliency of solar electricity.

14 SOLAR ENERGY↗
DOE OSTI2026

Automating Detection and Diagnosis of Faults, Failures, and Underperformance in PV Plants

The project developed hybrid physics-based and machine-learning methods for near-real-time detection of balance-of-system faults (e.g., string, combiner, and tracker outages) in utility-scale Photovoltaic plants, achieving over 50% true positive rates with under 10% false positives and significantly reducing engineering setup time. In the extended phase, the scope expanded to plant-level underperformance analysis and industry benchmarking through the SUPER.epri.com platform. SUPER standardizes data processing and performance metrics across more than 9 GWac and 120+ plants, enabling robust comparisons and insights into loss rates, inverter downtime, and capacity degradation.

14 SOLAR ENERGY↗
DOE OSTI2025

Planning Amidst Uncertainty: Identifying Core CCS Infrastructure Robust to Storage Uncertainty

Carbon Capture and Storage (CCS) is a critical technology for reducing anthropogenic CO2 emissions, but its large-scale deployment is complicated by uncertainties in geological storage performance. These uncertainties pose significant financial and operational risks, as underperforming storage sites can lead to costly infrastructure modifications, inefficient pipeline routing, and economic shortfalls. To address this challenge, we propose a novel optimization workflow that is based on mixed-integer linear programming and explicitly integrates probabilistic modeling of storage uncertainty into CCS infrastructure design. This workflow generates multiple infrastructure scenarios by sampling storage capacity distributions, optimally solving each scenario using a mixed-integer linear programming model, and aggregating results into a heatmap to identify core infrastructure components that have a low likelihood of underperforming. A risk index parameter is introduced to balance trade-offs between cost, CO2 processing capacity, and risk of underperformance, allowing stakeholders to quantify and mitigate uncertainty in CCS planning. Applying this workflow to a CCS dataset from the US Department of Energy’s Carbon Utilization and Storage Partnership project reveals key insights into infrastructure resilience. Reducing the risk index from 15% to 0% is observed to lead to an 83.7% reduction in CO2 processing capacity and a 77.1% decrease in project profit, quantifying the trade-off between risk tolerance and project performance. Furthermore, our results highlight critical breakpoints, where small adjustments in the risk index produce disproportionate shifts in infrastructure performance, providing actionable guidance for decision-makers. Unlike prior approaches that aimed to cheaply repair underperforming infrastructure, our workflow constructs robust CCS networks from the ground up, ensuring cost-effective infrastructure under storage uncertainty. These findings demonstrate the practical relevance of incorporating uncertainty-aware optimization into CCS planning, equipping decision-makers with a tool to make informed project planning decisions.

Olson, Daniel↗
DOE OSTI2023

A multiscale landscape approach for prioritizing river and stream protection and restoration actions

River and stream conservation programs have historically focused on a single spatial scale, for example, a watershed or stream site. Recently, the use of landscape information (e.g., land use and land cover) at multiple spatial scales and over large spatial extents has highlighted the importance of incorporating a landscape perspective into stream protection and restoration activities. Previously, we developed a novel framework that links information about watershed-, catchment-, and reach-scale integrity with stream biological condition using scatterplots and a landscape integrity map. Here we examined an application of this approach for streams in urban and other settings in King County, Washington State, United States, where we related stream macroinvertebrate condition to two indices of landscape integrity, the US Environmental Protection Agency's (USEPA) nationally available Index of Watershed Integrity (IWI) and Index of Catchment Integrity (ICI). We generated a scatterplot of IWI versus ICI for sample sites, where points represented site macroinvertebrate condition from poor to good. The same data were also visualized as a landscape integrity map that displayed catchments of King County according to the level of watershed and catchment integrity (high or low IWI/ICI). Almost three-quarters of poor-condition sites were associated with high-integrity watersheds and catchments (i.e., underperforming sites), which suggested that either one or both national indicators were insufficient for this area, and that sites underperformed because of local-scale factors. In response, we used a catchment-scale indicator related to forest condition (PctForestCat) after examining several GIS-based dispersal indicators from the National Hydrography Dataset and other candidates from the USEPA's StreamCat dataset. We then compared the results of the scatterplots and maps based on the current and original analyses and found that many of the sites previously classified as underperforming now performed as expected, that is, they were poor-condition sites in poor-condition catchments. This analysis demonstrates how results based on a national dataset can be improved by developing an alternative that represents regionally important stressors. The methods used to develop an effective landscape indicator based on StreamCat datasets, and the utility of the multiscale approach, could provide important tools for prioritizing, optimizing, and communicating stream conservation actions.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Wind Plant Performance Prediction Benchmark Phase 1 (Technical Report)

Financial risk resulting from the uncertainty associated with developing, owning, and operating wind power plants remains a barrier to reducing the levelized cost of energy (LCOE). On average, modern wind power plants in the U.S. underperform their expected annual energy output by 3.5-4.5% , with many underperforming by over 10%. To compensate for this uncertainty, investors require a larger return on investment (ROI) and apply "knock-down" factors that mask much of the underlying sources of uncertainty. Wind energy projects thus have reduced access to low-cost capital. Furthermore, operating wind plants often take a simple approach to estimating operations & maintenance (O&M) costs (e.g. straight-line estimates based on similar plants), which can eat into profits. To overcome these issues, the wind industry must improve the models they use for estimating wind plant performance and operations. An industry consortium (IC) requested that the National Renewable Energy Laboratory (NREL) lead a Department of Energy (DOE) working group to benchmark the accuracy of wind power plant energy predictions against real operational data. The IC was also motivated by DOE and NREL's potential to characterize systematic energy underperformance, identify sources of uncertainty, and explore root causes. The Wind Plant Performance Prediction (WP3) project was created out of this request, and this report represents the successful completion of Phase 1 of the WP3 project. During the project, wind plant owners provided both pre-construction and operational data to NREL. The pre-construction data was provided to wind resource assessment (WRA) consultants so they could conduct energy yield assessments (EYA). NREL took all of the completed EYAs, along with the operational data, and conducted an operational assessment to benchmark the EYA results against actual operational data. Given the large amounts of sensitive data required for this effort, as well as historical opposition to sharing data within industry, successful completion of Phase 1 represents an unprecedented milestone for industry data sharing. To improve the accuracy and confidence of pre-construction EYAs, wind plant owners and investors need better, more certain, energy yield predictions. The WP3 Benchmark Project is an industry-driven response to this reality. For the first time, industry has taken the important step of working together at scale, sharing valuable operational data with DOE and NREL in order to investigate the sources of bias and uncertainty in these energy estimates. This IC provides wind plant preconstruction and operational data to NREL in an organized and documented fashion and provides guidance and feedback as needed. The IC also provides introspection of the design of experiment, key metrics of success, data challenges, analysis best practices, and quality of results.

17 WIND ENERGY↗
DOE OSTI2023

Perspective: Performance Loss Rate in Photovoltaic Systems

Photovoltaic systems may underperform expectations for several reasons, including inaccurate initial estimates, suboptimal operations and maintenance, or component degradation. Accurate assessment of these loss factors aids in addressing root causes of underperformance and in realizing accurate expectations and models. The performance loss rate (PLR) is a commonly cited high‐level metric for the change in system output over time, but there is no precise, standard definition. Herein, an annualized definition of PLR that is inclusive of all loss factors and that can capture nonlinear changes to performance over time is proposed. The importance of distinguishing between recoverable and nonrecoverable losses which underly PLR is highlighted.

14 SOLAR ENERGY↗
DOE OSTI2022

Failure diagnosis and trend‐based performance losses routines for the detection and classification of incidents in large‐scale photovoltaic systems

Abstract Fault detection and classification in photovoltaic (PV) systems through real‐time monitoring is a fundamental task that ensures quality of operation and significantly improves the performance and reliability of operating systems. Different statistical and comparative approaches have already been proposed in the literature for fault detection; however, accurate classification of fault and loss incidents based on PV performance time series remains a key challenge. Failure diagnosis and trend‐based performance loss routines were developed in this work for detecting PV underperformance and accurately identifying the different fault types and loss mechanisms. The proposed routines focus mainly on the differentiation of failures (e.g., inverter faults) from irreversible (e.g., degradation) and reversible (e.g., snow and soiling) performance loss factors based on statistical analysis. The proposed routines were benchmarked using historical inverter data obtained from a 1.8 MWp PV power plant. The results demonstrated the effectiveness of the routines for detecting failures and loss mechanisms and the capability of the pipeline for distinguishing underperformance issues using anomaly detection and change‐point (CP) models. Finally, a CP model was used to extract significant changes in time series data, to detect soiling and cleaning events and to estimate both the performance loss and degradation rates of fielded PV systems.

14 SOLAR ENERGY↗
DOE OSTI2023

Voltage mapping and local defects identification in solar cells using non-contact method

Electroluminescence, infrared imaging, and current–voltage curve techniques are used to detect and map underperforming cells in a photovoltaic module or the modules in a photovoltaic string. In this work, we present a non-contact electrostatic voltmeter technique to detect and map the underperforming spots in a cell and the cells in a module. This non-contact technique directly maps the charged surface voltage (671mV) of the superstrate glass, and it had an 11-mV difference with the voltmeter values (660mV). Another data set of voltage values obtained by electroluminescence images conversion into a voltage map showed a difference of 7mV with the non-contact voltmeter values. Further, the direct voltage values obtained at various good and poor-performing spots of the cells using this technique are 3.69V and 3.79V and are validated using the voltage values obtained in electroluminescence analysis and the difference in voltage obtained by the two techniques is determined to be less than 2%. In this work, we combine the strengths of two complementary techniques of the electrostatic voltmeter (strength: quantitative) and electroluminescence (strength: spatial mapping) to obtain a quantitative spatial mapping of defects. Furthermore, this work is extendable to detect the poor-performing modules in solar PV power plants.

14 SOLAR ENERGY↗
DOE OSTI2022

Use of Physics to Improve Solar Forecast: Part II, Machine Learning and Model Interpretability

Machine learning (ML) models have been applied to forecast solar energy; however, they often lack clarity of interpretability and underlying physics. This work addresses such challenges by developing a hierarchy of ML models that gradually introduce predictors to improve the forecast accuracy based on a physics-based framework. Three ML models (ARIMA, LSTM, and XGBoost) are examined and compared with four physics-informed persistence models reported in Part I and the simple persistence model to assess the improvement of different models. The 7-year measurements at the U.S. Department of Energy's Atmospheric Radiation Measurement's Southern Great Plains Central Facility site are used for forecasts and evaluations. The results reveal that the step-by-step introduction of predictors leads to different improvements for models at different hierarchical levels. Comparison of the ML models with persistence models shows that LSTM and XGBoost outperform all the persistence models, with LSTM having the overall best performance; however, ARIMA underperforms the four physics-informed persistence models. This study demonstrates the importance and utility of incorporating physics into ML models in improving forecast accuracy by introducing a hierarchy of physics-based predictors, distinguishing predictor contributions, and enhancing the ML interpretability. The combined use of Global Horizontal Irradiance (GHI) and Direct Normal Irradiance (DNI) significantly improves the forecast accuracy compared to using individual irradiances alone because the pair contains more information on cloud-radiation interactions.

interpretability↗
DOE OSTI2021

PV Fleet Performance Data Initiative: Performance Index-Based Analysis

In this analysis, we report on 250 PV systems throughout the US, comprising 157 MWdc of system capacity and over 10,000 monthly PI values based on high-frequency (subhourly) energy data and satellite-based resource data. The distribution of PI values is analyzed, and multiple causes of underperformance are assessed, including first-year startup issues, snowfall and inverter downtime. An initial distribution of raw monthly PI values was collected with mean measured / modeled performance of PI = 0.935. After correcting for the three identifiable loss factors mentioned above, an average monthly performance of PI = 0.994 was obtained, with a distribution closely following a Gumbel Extreme Value distribution. In particular, inverter availability was found to contribute a system energy loss of 2.3% on average, except in the first six months of operation when availability losses are closer to 8%. Other startup issues beyond inverter downtime such as partial string outage contributed an additional 5% underperformance in the first year of operation across the fleet. Winter performance was also found to be 5%-10% below summer performance on average, likely due to snowfall. A simple linear relationship was found between snow loss and monthly snowfall accumulation in cm, indicating between 6% - 40% loss in months with nonzero snowfall, depending on climate.

14 SOLAR ENERGY↗
DOE OSTI2022

Plant-level performance and degradation of 31 GW DC of utility-scale PV in the United States

In this updated study, which samples 50% more capacity than the original and adds two additional years of operating history, we assess the performance of a fleet of 631 utility-scale PV plants totaling 31.0 GW DC (23.6 GW AC ) of capacity that achieved commercial operations in the United States from 2007-2018 and that have operated for at least two full calendar years. We use detailed information on individual plant characteristics, in conjunction with modeled irradiance data, to model expected or “ideal” capacity factors in each full calendar year of each plant’s operating history. A comparison of ideal versus actual first-year capacity factors finds that this fleet has modestly underperformed initial expectations (as modeled) on average, though perhaps due as much to modeling issues as to actual underperformance. We then analyze fleet-wide performance degradation in subsequent years by employing a “fixed effects” regression model to statistically isolate the impact of age on plant performance. The resulting average fleet-wide degradation rate of -1.2%/year (±0.1%) represents a slight improvement (seemingly driven by the oldest plants in our sample) over the -1.3%/year (±0.2%) found in our original study, yet is still of greater magnitude than is commonly found. We emphasize, however, that these fleet-wide estimates reflect both recoverable and unrecoverable degradation across the entire plant, and so will naturally be of greater magnitude than module- or cell-level studies, and/or studies that focus only on unrecoverable degradation. Moreover, when focusing on a sub-sample of newer and larger plants with higher DC:AC ratios—i.e., plants that more-closely resemble what is being built today—we find a more moderate sample-wide average performance decline of -0.7%/year (±0.4%), which is more in line with other estimates from the recent literature.

14 SOLAR ENERGY↗
DOE OSTI2023

Optimizing Performance on Trinity Utilizing Machine Learning, Proxy Applications and Scheduling Priorities

The sheer number of nodes continues to increase in today’s supercomputers, the first half of Trinity alone contains more than 9400 compute nodes. Since the speed of today’s clusters are limited by the slowest nodes, it more important than ever to identify slow nodes, improve their performance if it can be done, and assure minimal usage of slower nodes during performance critical runs. This is an ongoing maintenance task that occurs on a regular basis and, therefore, it is important to minimize the impact upon its users by assessing and addressing slow performing nodes and mitigating their consequences while minimizing down time. These issues can be solved, in large part, through a systematic application of fast running hardware assessment tests, the application of Machine Learning, and making use of performance data to increase efficiency of large clusters. Proxy applications utilizing both MPI and OpenMP were developed to produce data as a substitute for long runtime applications to evaluate node performance. Machine learning is applied to identify underperforming nodes, and policies are being discussed to both minimize the impact of underperforming nodes and increase the efficiency of the system. In this paper, I will describe the process used to produce quickly performing proxy tests, consider various methods to isolate the outliers, and produce ordered lists for use in scheduling to accomplish this task.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Synergy of Ion Doping and Spiral Array Architecture on Ti 2 Nb 10 O 29 : A New Way to Achieve High-Power Electrodes

Ameliorating electronic/ionic transport and structural stability of electrode materials is important to the development of power-intensive lithium ion batteries. Despite its great potential as a high-power anode, titanium niobium oxide (Ti 2 Nb 10 O 29 , TNO) still underperforms due to its unsatisfactory electronic/ionic conductivity. In this work, a powerful synergistic strategy by combining ion doping and spiral array architecture to boost high-rate performance of TNO is reported. Cr 3+ doped TNO nanoparticles (Cr-TNO) of 5-10 nm intimately grow on a conductive vertical graphene@TiC-C (VGTC) skeleton, forming novel Cr-TNO@VGTC spiral arrays. The unique spiral growth of TNO is achieved due to the confinement effect of VGTC skeleton. Meanwhile, a more open TNO crystal structure with faster ion transfer paths and enhanced structural stability is realized by Cr 3+ doping, demonstrated via density functional theory calculation and in situ synchrotron X-ray diffraction technique. Benefiting from the superior conductive network, enhanced intrinsic electronic/ionic conductivity of Cr-TNO and reinforced structural stability, the Cr-TNO@VTC arrays show prominent high-power performance with a large capacity of 220 mAh g -1 at 40 C (power density of ≈11 kW kg -1 ) and superior durability (91% retention after 500 cycles). This work provides a new path for the construction of widespread high-power electrodes for fast energy storage.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Halide Perovskites and Their Derivatives for Efficient, High‐Resolution Direct Radiation Detection: Design Strategies and Applications

Abstract The past decade has witnessed a rapid rise in the performance of optoelectronic devices based on lead‐halide perovskites (LHPs). The large mobility‐lifetime products and defect tolerance of these materials, essential for optoelectronics, also make them well‐suited for radiation detectors, especially given the heavy elements present, which is essential for strong X‐ray and γ‐ray attenuation. Over the past decade, LHP thick films, wafers, and single crystals have given rise to direct radiation detectors that have outperformed incumbent technologies in terms of sensitivity (reported values up to 3.5 × 10 6 µC Gy air −1 cm −2 ), limit of detection (directly measured values down to 1.5 nGy air s −1 ), along with competitive energy and imaging resolution at room temperature. At the same time, lead‐free perovskite‐inspired materials (e.g., methylammonium bismuth iodide), which have underperformed in solar cells, have recently matched and, in some areas (e.g., in polarization stability), surpassed the performance of LHP detectors. These advances open up opportunities to achieve devices for safer medical imaging, as well as more effective non‐invasive analysis for security, nuclear safety, or product inspection applications. Herein, the principles behind the rapid rises in performance of LHP and perovskite‐inspired material detectors, and how their properties and performance link with critical applications in non‐invasive diagnostics are discussed. The key strategies to engineer the performance of these materials, and the important challenges to overcome to commercialize these new technologies are also discussed.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) peak intensity normalization for complex mixture analyses

Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) is a preferred technique for analyzing complex organic mixtures. Currently, there is no consensus normalization approach, nor an objective method for selecting one, for quantitative analyses of FT-ICR-MS data. We investigate a method to evaluate and score the amount of bias various normalization approaches introduce into the data. We evaluate the ability of the Statistical Procedure for the Analysis of Normalization Strategies (SPANS) to guide selection of appropriate normalization approaches for two different FT-ICR-MS datasets. Further, we test the robustness of SPANS results to changes in SPANS parameter values and assess the impact of using various normalization approaches on downstream statistical analyses. The normalization approach identified by SPANS differed for the two datasets. Normalization methods impacted the statistical significance of peaks differently, underscoring the importance of carefully evaluating potential methods. More consistent SPANS scores resulted when at least 120 significant peaks are used, where larger sets of peaks were obtained by increasing the p-value threshold. Interestingly, we show that Total Sum Scaling and highest peak normalization, used in previous studies, underperformed relative to SPANS-recommended normalization approaches. While there is no single, best normalization method for all datasets, SPANS provides a mechanism to identify an appropriate normalization method for analyzing FT-ICR-MS data quantitatively. As a result, the number of peaks used in the background distributions of SPANS contributes more significantly to the reproducibility of results than the p-value thresholds used to obtain those peaks.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Extraordinarily strong magneto-responsiveness in phase-separated LaFe 2 Si

Materials responding vigorously to minor variations of external stimuli with negligible hysteresis could revolutionize many of the energy technologies, including refrigeration, actuation, and sensing. Here, we report a combined experimental and theoretical study of a two-phase composite, naturally formed at the LaFe 2 Si stoichiometry, which exhibits a nearly anhysteretic, two-step first-order ferromagnetic-to-paramagnetic phase transformation with enhanced sensitivity to an external magnetic field. Other unusual properties include a large plateau-like positive magnetoresistance, magnetic-field-induced temperature and entropy changes occurring over a wide temperature range, and a Griffiths-like phase associated with short-range ferromagnetic clustering in the paramagnetic state. The heat capacity, magnetization, Mössbauer spectroscopy, and electrical resistivity, all exhibit characteristic, unusually sharp, first-order discontinuities even in magnetic fields as high as 100 kOe. We expect that similar phenomena could be designed in other mixed-phase systems, leading to novel functionalities, such as giant caloric effects in many yet undiscovered or/and underperforming intermetallic compounds.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Assessing geothermal/solar hybridization – Integrating a solar thermal topping cycle into a geothermal bottoming cycle with energy storage

Geothermal and concentrating solar power (CSP) technologies typically use heat at different temperatures in their commercial deployments. This enables a technically viable hybridization of a solar topping cycle and a geothermal bottoming cycle at locations where both resources are available. In this article, an underperforming geothermal power system based at Burley, Idaho is used as a baseline to investigate the technical and economic potential of such a hybrid cycle. A direct thermal energy storage (TES) system is also integrated to overcome the intermittency of the solar resource. Design and off-design behavior of key components are modelled to simulate the annual performance of the hybrid system on an hourly basis. The sizing of the solar field and thermal storage is investigated and is seen to significantly impact the annual electricity generation and efficiency. Various cost scenarios for the solar field and thermal energy storage are investigated. An economic metric - levelized cost of electricity (LCOE) - is used to optimize the solar field sizing and TES capacity. The hybrid plant converts the additional solar heat input into additional work with an efficiency of 32.9%. Retrofitting a geothermal plant with solar and eight hours of energy storage can achieve an LCOE of 0.136 $/kWhe in current cost scenarios and 0.081 $/kWhe in a future cost reduction scenario. Although the hybrid system cannot directly compete with current PV systems without batteries, it has an LCOE 32% lower than that of a PV-battery system. This paper provides insights into the research and development of the future grid with a high renewable energy penetration and encourages further study of energy hybridization for improved efficiencies and economics.

15 GEOTHERMAL ENERGY↗
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↗