Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Explanability”

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 613 records · Page 34
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Response to a comment on Dessert et al. “The dark matter interpretation of the 3.5 keV line is inconsistent with blank-sky observations”

The dark matter explanation of the 3.5 keV line is strongly disfavored by our work in Dessert et al. (2020). Boyarsky et al. (2004) questions that conclusion: modeling additional background lines is claimed to weaken the limit sufficiently to re-allow a dark matter interpretation. Here, we respond as follows. (1) A more conservative limit is obtained by modeling additional lines; this point appeared in its entirety in our work in Dessert et al. (2020), though we also showed that the inclusion of such lines is not necessary. (2) Despite suggestions in Boyarsky et al. (2004), even the more conservative limits strongly disfavor a decaying dark matter origin of the 3.5 keV line.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Missing in axion: Where are XENON1T’s big black holes?

We pioneer the black hole mass gap as a powerful new tool for constraining new particles. A new particle that couples to the Standard Model – such as an axion – acts as an additional source of loss in the cores of population-III stars, suppressing mass lost due to winds and quenching the pair-instability. This results in heavier astrophysical black holes. As an example, using stellar simulations we show that the solar axion explanation of the recent XENON1T excess implies astrophysical black holes of ~56M ⊙ , squarely within the black hole mass gap predicted by the Standard Model.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

Surrogate model evaluation and building energy benchmarking for commercial buildings

Building energy consumption benchmarking involves challenges associated with various energy patterns for different building types; heating, ventilating, and air-conditioning (HVAC) system types; and climates. Given significant variation in energy use patterns, accurate prediction of long-term energy use using surrogate models remains challenging. Multiple linear regression (MLR) is commonly used for building energy benchmarking because of its simple structure; however, it lacks accuracy compared to other black-box models. Although many studies have compared surrogate models and offer guidance on model selection based on metrics, they do not provide detailed analysis on improving the surrogate model accuracy. In this paper, we implement a surrogate model using polynomial ridge regression (i.e., MLR with interaction terms combined with ridge regularization) for small office and retail strip mall buildings across six HVAC system types and all climate zones, for electricity and natural gas in baseline and proposed scenarios. A simulation workflow is developed using OpenStudio TM /EnergyPlus TM to generate simulation data using measures over a wide range of efficiency inputs. Enhancements based on statistical insights are used for improving the model accuracy using filters, input transformations, and change points. Surrogate models achieved average coefficient of variation of the root mean squared error (CVRMSE) values of 2.17, 1.06, 2.05, and 3.26 for proposed electricity, proposed natural gas, baseline electricity, and baseline natural gas, respectively, with enhancements reducing CVRMSE by an average of 14.9% across all combinations. We provide model interpretation via Shapley additive explanations to determine which input variables most influence energy consumption and provide supportive arguments for enhancements.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Explainable machine learning model for multi-step forecasting of reservoir inflow with uncertainty quantification

We propose an explainable machine learning (ML) model with uncertainty quantification (UQ) to improve multi-step reservoir inflow forecasting. Traditional ML methods have challenges in forecasting inflows multiple days ahead, and lack explainability and UQ. To address these limitations, we introduce an encoder–decoder long short-term memory (ED-LSTM) network for multi-step forecasting, employ the SHapley Additive exPlanation (SHAP) technique for understanding the influence of hydrometeorological factors on inflow prediction, and develop a novel UQ method for prediction trustworthiness. We apply these methods to forecast 7-day inflow in snow-dominant and rain-driven reservoirs. The results demonstrate the effectiveness of the ED-LSTM model, with high forecasting accuracy for short lead times. Our UQ method provides reliable uncertainty estimates, covering 90% of data with a 90% confidence level. The SHAP analysis reveals the importance of historical inflow and precipitation as influential factors. These findings and methods may support reservoir operators in optimizing water resources management decisions.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Sound velocity and compressibility of melts along the hedenbergite (CaFeSi2O6)-diopside (CaMgSi2O6) join at high pressure : Implications for stability and seismic signature of Fe-rich melts in the mantle

Iron-rich silicate melts play an important role in the magmatic history of the Earth and the 16 Moon. However, their elastic properties at high pressures, especially the sound velocity, are poorly 17 understood. Here we determined the ultrasonic sound velocity for the first time of a hedenbergite 18 (Hd, CaFeSi2O6) melt and a melt mixture of 50 mol% hedenbergite + 50 mol% diopside (Hd50Di50) 19 at high pressure and temperature conditions up to 6 GPa and 2329 K, using high-pressure ultrasonic 20 technique combined with synchrotron radiation in a multi-anvil apparatus. Our results show that 21 Fe can significantly reduce the sound velocity while increase the density of silicate melts. 22 Comparing the Di, Hd, and Hd50Di50 melts, we find that the sound velocity does not mix linearly 23 2 for melts in the Hd-Di join, whereas the density for Hd-Di melts at high pressures can be well-24 described by linear mixing. Combined with melt geometry and melt compaction models, we 25 applied our results to study the stability and seismic signature of Fe-rich silicate melts in the Earth’s 26 upper mantle. For the low-velocity zone (LVZ) in mantle asthenosphere, although the degree of 27 seismic velocity reduction can be explained by the presence of a small amount of partial melt 28 distributed in film/band geometry along grain boundaries, Fe-rich melts formed at this depth are 29 unlikely to be gravitationally stable, but may be dynamically unextractable if the melt supply is 30 continuous. For the low-velocity layer (LVL) above the mantle transition zone, the presence of Fe-31 rich melts (with FeO>~10 wt%) distributed in textural equilibrium with the ambient mantle is a 32 plausible explanation.

diopside↗
DOE OSTI2024

Bioavailability of molybdenite to support nitrogen fixation on early Earth by an anoxygenic phototroph

Biological nitrogen fixation, which converts atmospheric dinitrogen to ammonia, is catalyzed mostly by Mo-nitrogenase and is a primary contributor to bioavailable nitrogen on early Earth. Mo-nitrogenase is believed to have evolved during the Archean, despite the extremely low concentration of dissolved Mo. However, it remains unclear whether Mo minerals could serve as a source of Mo to support the prevalence of Mo-nitrogenase on early Earth. Here, in this study, we investigated the bioavailability of molybdenite by incubating it with a metabolically ancient anoxygenic phototroph (Rhodopseudomonas palustris) under anoxic conditions. In the laboratory, R. palustris utilized molybdenum from molybdenite as a cofactor for nitrogen fixation. This bacterium extracted Mo from molybdenite by secreting molybdophores rhodopetrobactin A and B and by expressing Mo transport proteins. Surface-sensitive techniques demonstrated significant changes in surface chemistry of molybdenite after its interaction with cells. These findings provide novel explanations for the prevalence of Mo-nitrogenase on early Earth, with significant implications for nitrogen fixation in modern Mo-deficient environments.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Enhanced Oblique Decision Tree Enabled Policy Extraction for Deep Reinforcement Learning in Power System Emergency Control

Deep reinforcement learning (DRL) algorithms have successfully solved many challenging problems in various power system control scenarios. However, their decision-making process is usually regarded as black-boxes. Furthermore, how DRL models interact with human intelligence remains an open problem. Thus, this paper proposes a policy extraction framework to extract a complex DRL model into an explainable policy. This framework includes three parts: 1) DRL training and data generation. We train an agent for a specific control task and generate data, which contains the control policy of the agent. 2) Policy extraction. We propose an information gain rate based weighted oblique decision tree (IGR-WODT) for DRL policy extraction. 3) Policy evaluation. We define three metrics to evaluate the performance of the proposed approach. A case study for the under-voltage load shedding problem shows that the IGR-WODT presents a performance enhancement compared with DRL, weighted oblique decision tree, and univariate decision tree. The proposed policy extraction method could provide an intuitive explanation of the neural network decision-making process to the dispatchers when making final decisions on power grid operation. Also, the resulted rule-based controller could replace the deep neural network-based controller in many field edge devices with limited computing resources, providing comparable performance.

deep reinforcement learning↗
DOE OSTI2025

Computationally efficient models for aqueous organic redox flow batteries

The rising usage of intermittent energy has garnered the need for large scale energy storage systems. Redox flow batteries (RFB) based energy storage system shows promising potential. Numerical simulations and machine learning approaches have been widely used to study RFB performance. The development of autonomous material discovery framework and digital twin of energy storage system usually needs to query cell performance through fast response models. In this study, two computationally efficient models are introduced: a physics-based analytical flow battery model (EZBattery), and a machine learning operator model (Deep Operator Network, denoted by DeepONet). Both models can provide cell performance near instantly, and prediction accuracy was systematically examined on an application of evaluating the performances of a 780 cm 2 aqueous organic redox flow battery (AORFB), using potential anolyte candidates in dihydroxyphenazine (DHP)-based family of organic materials. A validated computationally expansive 3-dimensional multi-physics finite element model by COMSOL was used as the ground truth and provided the training data set for the DeepONet. 1280 samples were generated with 10 properties to mimic the different possible anolyte candidates, and the cell performances were evaluated under 10 different combined operating conditions. The accuracy comparisons for the two computationally efficient models show that both models can provide comparable accuracy in predicting cell charging/discharging voltage curves. DeepONet can provide slightly higher overall accuracy than EZBattery with faster calculation speed, but highly relies on the training dataset. EZBattery does not need a training dataset and can provide interpretable physics-based explanations of the results, while being more flexible to adjust to adapt any different cell designs, flow battery architectures, and electrolyte materials.

Analytical model↗
DOE OSTI2023

Improving the accuracy of freight mode choice models: A case study using the 2017 CFS PUF data set and ensemble learning techniques

Here, the US Census Bureau has collected two rounds of experimental data from the Commodity Flow Survey, providing shipment-level characteristics of nationwide commodity movements, published in 2012 (i.e., Public Use Microdata) and in 2017 (i.e., Public Use File). With this information, data-driven methods have become increasingly valuable for understanding detailed patterns in freight logistics. In this study, we used the 2017 Commodity Flow Survey Public Use File data set to explore building a high-performance freight mode choice model, considering three main improvements: (1) constructing local models for each separate commodity/industry category; (2) extracting useful geographical features, particularly the derived distance of each freight mode between origin/destination zones; and (3) applying additional ensemble learning methods such as stacking or voting to combine results from local and unified models for improved performance. The proposed method achieved over 92% accuracy without incorporating external information, an over 19% increase compared to directly fitting Random Forests models over 10,000 samples. Furthermore, SHAP (Shapely Additive Explanations) values were computed to explain the outputs and major patterns obtained from the proposed model. The model framework could enhance the performance and interpretability of existing freight mode choice models.

42 ENGINEERING↗
DOE OSTI2025

A methodology for corrosion testing ODS steels in liquid tin under reactor conditions in HFIR

Liquid metal–based divertor concepts have promising attributes for application in fusion reactors. Liquid tin is being considered as one of those liquid metals due to its excellent thermophysical properties. However, tin is extremely corrosive to steels at elevated temperatures, and the coupled effects of corrosion with neutron irradiation for application in fusion reactors have not been quantified. Researchers at Oak Ridge National Laboratory (ORNL) have designed and irradiated a series of capsules in the High Flux Isotope Reactor (HFIR) to help address gaps in the literature with respect to coupled corrosion and neutron irradiation effects. The capsules are filled with solid tin designed to melt due to the gamma heating in HFIR and allow interaction with the specimens. The capsules contain a SiC thermometer for post-irradiation temperature verification. Five capsules were irradiated in HFIR for 10.5 days, accumulating 0.93 dpa in the FeCrAl specimens. Post-irradiation temperature verification was performed on the SiC thermometers, and temperatures were observed to be higher than what was expected from thermal models. An explanation for the model underprediction is given herein. In conclusion, the capsules described within this article are the first to irradiate liquid metal within the HFIR flux trap and demonstrated the safety basis for continued liquid coolant studies in HFIR.

FeCrAl↗
DOE OSTI2020

Sulfur isotope fractionation between aqueous and carbonate-associated sulfate in abiotic calcite and aragonite

Sulfate (SO 4 2– ) incorporated into calcium carbonate minerals enables measurements of sulfur (S) isotope ratios in carbonate rocks. This Carbonate Associated Sulfate (CAS) in marine carbonate minerals is thought to faithfully represent the S isotope composition of the seawater sulfate incorporated into the mineral, with little or no S isotope fractionation in the process. However, comparison between different calcifying species reveals both positive and negative S isotope fractionation between CAS and seawater sulfate, and a large range of S isotope ratios can be found within a single rock sample, depending on the component measured. To better understand the isotopic effects associated with sulfate incorporation into carbonate minerals, we precipitated inorganic calcite and aragonite over a range covering more than two orders of magnitude of sulfate concentration and precipitation rate. Coupled measurements of CAS concentration, S isotope composition and X-ray absorption near-edge spectra (XANES) permit characterization and explanation of the observed dependence of S isotope fractionation between CAS and aqueous sulfate (CAS-SO 4 2– isotope fractionation) on sulfate concentration and precipitation rate. In aragonite, the CAS-SO 4 2– isotope fractionation is 1.0 ± 0.3‰ and independent of the sulfate (and CAS) concentration. In contrast, we note the CAS-SO 4 2– isotope fractionation in calcite covaries strongly with the sulfate concentration and weakly with the precipitation rate, between values of 1.3 ± 0.1 and 3.1 ± 0.6‰. We suggest that the correlation between aqueous sulfate concentration and CAS-SO 4 2– isotope fractionation in calcite reflects a dependence of the equilibrium S isotope fractionation on the concentration of CAS, through the effect of the sulfate impurity on the carbonate mineral’s energetic state.

58 GEOSCIENCES↗
DOE OSTI2024

Soft interface instability and gas flow channeling in low-permeability deformable media

Understanding gas percolation through a clay layer or a shale formation is of great importance for the development of a geologic repository for nuclear waste disposal, a subsurface system for gas storage, and an engineering approach for hydrocarbon extraction from unconventional reservoirs. Gas injection experiments have revealed complex dynamic behaviours of gas percolation through water saturated compacted bentonite, characterized by a high breakthrough pressure, rapid breakthrough, a pressure/stress decay after the breakthrough, a relatively high migration rate, high-frequency periodic/nonperiodic variations in flow rate, stepwise rate reductions during relaxation, and low gas saturation over the whole process, all indicating channelling nature of the processes. Using linear stability analyses, we show that this channelling can autonomously emerge from the instability of the deformable interface between the injected gas and the compacted bentonite matrix driven by local stress concentration, pore dilation, and hydrologic gradient. Channel patterns formed would possess a fractal geometry. We further show that, once a percolating channel is established, the gas injected would percolate through the channel in a chain of gas bubbles, also due to the interface instability, resulting in periodic/chaotic variations in gas flow rate. Our work provides a unified explanation for key features observed for gas percolation in low-permeability deformable media. The work also suggests a possibility of designing an engineered barrier system for a nuclear waste repository that can have controllable gas release while limit water transport.

Bentonite↗
DOE OSTI2021

Results from single-shock Marble experiments studying thermonuclear burn in the presence of heterogeneous mix on the National Ignition Facility

The Marble campaign on the National Ignition Facility investigates the effect of heterogeneous mix on thermonuclear burn for comparison to a probability distribution function (PDF) burn model. Marble utilizes plastic capsules filled with deuterated plastic foam and a fill gas containing tritium. As the capsules implode, the deuterium in the foam mixes with the tritium gas, and DT neutrons are produced as the shocks compress and heat the mixture. The yield of DT neutrons is dependent on the uniformity of the mix, with more heterogeneous mix producing fewer neutrons. In Marble, the heterogeneity of the mix is controlled by varying the diameter of voids introduced into the foam. The first NIF Marble campaign has been executed in which the Marble capsules were indirectly driven with a single strong shock using NIF hohlraums. The experiments produce a low-convergence, high-ion-temperature implosion. The ratio of DT to DD neutron yield is largely consistent with uniform atomic mix for fine-pore foam, and increases slightly with void diameter, contrary to 1D simulations using the PDF burn model. Recent 3D high-resolution simulations of similar experiments performed on the Omega Laser Facility suggest an explanation.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2023

A machine learning approach for identifying variables associated with risk of developing neutralizing antidrug antibodies to factor VIII

A key unmet need in the management of hemophilia A (HA) is the lack of clinically validated markers that are associated with the development of neutralizing antibodies to Factor VIII (FVIII) (commonly referred to as inhibitors). This study aimed to identify relevant biomarkers for FVIII inhibition using Machine Learning (ML) and Explainable AI (XAI) using the My Life Our Future (MLOF) research repository. The dataset includes biologically relevant variables such as age, race, sex, ethnicity, and the variants in the F8 gene. In addition, we previously carried out Human Leukocyte Antigen Class II (HLA-II) typing on samples obtained from the MLOF repository. Using this information, we derived other patient-specific biologically and genetically important variables. These included identifying the number of foreign FVIII derived peptides, based on the alignment of the endogenous FVIII and infused drug sequences, and the foreign-peptide HLA-II molecule binding affinity calculated using NetMHCIIpan. The data were processed and trained with multiple ML classification models to identify the top performing models. The top performing model was then chosen to apply XAI via SHAP, (SHapley Additive exPlanations) to identify the variables critical for the prediction of FVIII inhibitor development in a hemophilia A patient. Using XAI we provide a robust and ranked identification of variables that could be predictive for developing inhibitors to FVIII drugs in hemophilia A patients. These variables could be validated as biomarkers and used in making clinical decisions and during drug development. The top five variables for predicting inhibitor development based on SHAP values are: (i) the baseline activity of the FVIII protein, (ii) mean affinity of all foreign peptides for HLA DRB 3, 4, & 5 alleles, (iii) mean affinity of all foreign peptides for HLA DRB1 alleles), (iv) the minimum affinity among all foreign peptides for HLA DRB1 alleles, and (v) F8 mutation type.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Advancing spatiotemporal forecasts of CO 2 plume migration using deep learning networks with transfer learning and interpretation analysis

Accurate and timely forecasts of CO 2 plume distribution throughout the injection and post-injection phases are crucial for detecting plume migration, assessing leakage risks, and supporting operational decisions in geologic carbon storage (GCS). Current convolutional neural network-based approaches primarily focus on spatial information and overlook temporal dependencies in plume distributions, thus limiting their ability to capture dynamic movement effects and provide accurate predictions of plume migration. In this work, we propose two deep learning models, Auto-Encoder (AE)-LSTM and Encoder-Decoder (ED)-ConvLSTM, each uniquely designed to capture both spatial and temporal features. We apply the proposed methods to forecast the dynamic distribution of CO 2 plumes based on 108 reservoir simulations over a 30-year injection and a 30-year post-injection period. The results indicate that the ED-ConvLSTM model outperforms the AE-LSTM model in accurately predicting the spatiotemporal dynamics of CO 2 plume migration, achieving R 2 values above 0.99. To provide a deeper understanding of these model predictions, we employ a gradient-based explanation method on the trained models. This approach provides insights into the influence of input variables on plume migration forecasts and uncovers the underlying prediction mechanisms of the proposed models. Furthermore, we introduce a transfer learning technique, enabling fast and accurate plume migration forecasting in the post-injection phase by leveraging the trained model during the injection phase. This reduces the necessity for extensive data collection or re-training. In conclusion, the methods proposed in our work enhances the performance and interpretability of CO 2 plume migration forecasts, thereby facilitating informed decision-making throughout the entire lifecycle of GCS applications.

58 GEOSCIENCES↗
DOE OSTI2025

Deep learning model for fast, science-based forecasting of fluid migration along faults in geologic carbon storage scenarios

Effective long-term geologic storage depends on robust site selection and credible, science-based forecasting of subsurface behavior to ensure storage integrity. For this work, we develop a deep learning–based reduced-order model (ROM) to quantify potential carbon dioxide (CO₂) and brine migration through geological faults. The ROM combines a Transformer model for binary classification and a Stacked Ensemble for regression, trained on a comprehensive dataset generated from 1400 physics-based reservoir simulations. Key geologic and operational parameters—including fault geometry, reservoir structure, and injection conditions—were systematically varied to capture a wide range of fluid migration scenarios. The ROM accurately predicts the onset of migration, cumulative migration volumes of both CO₂ and brine, and associated migration rates, as compared to an independent set of validation simulations, while significantly reducing computational cost compared to traditional simulation methods. Model performance was evaluated across diverse fault configurations, revealing that shallow reservoir geometry and fault angle are among the most influential factors governing migration behavior. Sensitivity analysis using SHapley Additive exPlanations (SHAP) provided interpretability, revealing distinct patterns in how geological and operational features drive transient versus cumulative migration outcomes. The ROM’s ability to rapidly simulate fault migration scenarios enables efficient sensitivity analyses, scenario evaluations, and decision support for site selection and monitoring design. This approach enhances the safety, scalability, and long-term operational performance of geologic carbon storage (GCS) systems by providing a robust, interpretable tool for predicting subsurface fluid migration and assessing fault-related migration potential.

42 ENGINEERING↗
DOE OSTI2020

Elucidating the Impact of Flow on Material-Sensitive Critical Heat Flux and Boiling Heat Transfer Coefficients: An Experimental Study with Various Materials

In this work, steady-state internal flow boiling critical heat flux (CHF) experiments were carried out at 10°C inlet subcooling and atmospheric pressure (84 kPa at an altitude of about 5300 feet) with mass flux ranging from 200 to 2,000 kg/m 2 -s on seven tube materials, including the accident-tolerant fuel cladding candidate FeCrAl (C26M) alloy. At a relatively low flow rate, the sensitivity of CHF to different materials was appreciable (the average difference among the tested materials was 19% at 200 kg/m 2 -s). As the absolute CHF differences stayed constant for higher flow rates, the increased flow rate led to reduced relative CHF differences among the tested materials (average relative difference of 15% at 1,000 kg/m 2 -s). FeCrAl alloy claddings that were oxidized under simulated Light Water Reactor (LWR) conditions yielded CHF comparable to as-received FeCrAl alloy cladding, with less than 1.8% difference. Experimental CHF results imply marginal material sensitivity impacts on the departure from nucleate boiling ratio in steady-state nuclear reactor cores. Yet, for accident progression analysis, material sensitivity may hold importance, as boiling behavior variation at low mass flux is appreciable. Macroscopic material parameters (surface wettability, surface roughness, and thermal properties) do not provide a suitable explanation for observed material-sensitive CHFs. Yet, surface roughness and thermal effusivity revealed power and linear relations, respectively, to the nucleate boiling heat transfer coefficients at low mass flux. At high mass flux, material sensitivity on nucleate boiling heat transfer significantly decreased due to the increased shear stress on boiling surface overwhelming material-related boiling implications.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Fouling modeling and prediction approach for heat exchangers using deep learning

In this article, we develop a generalized and scalable statistical model for accurate prediction of fouling resistance using commonly measured parameters of industrial heat exchangers. This prediction model is based on deep learning where a scalable algorithmic architecture learns non-linear functional relationships between a set of target and predictor variables from large number of training samples. Here, the efficacy of this modeling approach is demonstrated for predicting fouling in an analytically modeled cross-flow heat exchanger, designed for waste heat recovery from flue-gas using room temperature water. The performance results of the trained models demonstrate that the mean absolute prediction errors are under 10 –4 KW –1 for flue-gas side, water side and overall fouling resistances. The coefficients of determination (R 2 ), which characterize the goodness of fit between the predictions and observed data, are over 99%. Even under varying levels of measurement noise in the inputs, we demonstrate that predictions over an ensemble of multiple neural networks achieves better accuracy and robustness to noise. We find that the proposed deep-learning fouling prediction framework learns to follow heat exchanger flow and heat transfer physics, which we confirm using locally interpretable model agnostic explanations around randomly selected operating points. Overall, we provide a robust algorithmic framework for fouling prediction that can be generalized and scaled to various types of industrial heat exchangers.

42 ENGINEERING↗