Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ML T&E”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

A machine learning approach to fracture mechanics problems

Analytical and empirical solutions to engineering problems are usually preferred because of their convenience in applications. However, they are not always accessible in complex problems. A new class of solutions, based on machine learning (ML) models such as regression trees and neural networks (NNs), are proposed and their feasibility and value are demonstrated through the analysis of fracture toughness measurements. It is found that both solutions based on regression trees and NNs can provide accurate results for the specific problem, but NN-based solutions outperform regression-tree-based solutions in terms of their simplicity. Here this example demonstrates that ML solutions are a major improvement over analytical and empirical solutions in terms of both reliable functionality and rapid deployment. When analytical solutions are not available, the use of ML solutions can overcome the limitations of empirical solutions and substantially change the way that engineering problems are solved.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Overview of leakage scenarios in supervised machine learning

Machine learning (ML) provides powerful tools for predictive modeling. ML’s popularity stems from the promise of sample-level prediction with applications across a variety of fields from physics and marketing to healthcare. However, if not properly implemented and evaluated, ML pipelines may contain leakage typically resulting in overoptimistic performance estimates and failure to generalize to new data. This can have severe negative financial and societal implications. Our aim is to expand understanding associated with causes leading to leakage when designing, implementing, and evaluating ML pipelines. Illustrated by concrete examples, we provide a comprehensive overview and discussion of various types of leakage that may arise in ML pipelines.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

The Evaluation of Machine Learning Techniques for Isotope Identification Contextualized by Training and Testing Spectral Similarity

Precise gamma-ray spectral analysis is crucial in high-stakes applications, such as nuclear security. Research efforts toward implementing machine learning (ML) approaches for accurate analysis are limited by the resemblance of the training data to the testing scenarios. The underlying spectral shape of synthetic data may not perfectly reflect measured configurations, and measurement campaigns may be limited by resource constraints. Consequently, ML algorithms for isotope identification must maintain accurate classification performance under domain shifts between the training and testing data. To this end, four different classifiers (Ridge, Random Forest, Extreme Gradient Boosting, and Multilayer Perceptron) were trained on the same dataset and evaluated on twelve other datasets with varying standoff distances, shielding, and background configurations. A tailored statistical approach was introduced to quantify the similarity between the training and testing configurations, which was then related to the predictive performance. Wilcoxon signed-rank tests revealed that the OVR-wrapped XGB significantly outperformed the other algorithms, with confidence levels of 99.0% or above for the 133Ba, 60Co, 137Cs, and 152Eu sources. The findings from this work are significant as they outline techniques to promote the development of robust ML-based approaches for isotope identification.

domain adaptation↗
NASA NTRS1987

Space Shuttle inflight and postflight fluid shifts measured by leg volume changes

This is a study of the inflight and postflight leg volume changes associated with spaceflight on Space Shuttle missions. The results show an inflight volume loss of 2 l from the lower extremities, 1 l from each leg, representing an 11.6 percent volume change. The vast majority of this change appears to be a shift in body fluids, both intravascular and extravascular. The fluid shift occurs mostly on Mission Day One and is essentially complete by 6 to 10 hr. The regional origin of shift and leg volume changes shows a far greater absolute volume (708 ml vs. 318 ml) and percentage (69 percent vs. 31 percent) of the total change coming from the higher as compared to the lower leg. Postflight, the return of fluid to the lower extremities occurs rapidly with the majority of volume return complete within 1.5 hr postlanding. At 1 week postflight, there is a residual leg volume decrement of 283 ml or 3.2 percent that is probably due to tissue loss secondary to atrophic deconditioning and weight loss.

Moore, Thomas P.↗
NASA NTRS1990

Prepenetrant Etchant For Incology(R) 903 Weld Overlays

Etching solution developed for use prior to type-IVc penetrant inspection of Incology(R) 903 weld overlays. Formulated as follows: 80 g ferric chloride hexahydrate, 300 mL reagent-grade hydrochloric acid, 25 mL food- or reagent-grade phosphoric acid, and 100 mL ethylene glycol. Gives more reasonable range of etching time and reduces probability of overetching and resulting damage. Stored indefinitely.

O'Tousa, Joseph E.↗
DOE OSTI2023

Baltimore Ecosystem Study: Soil atmosphere fluxes of carbon dioxide, nitrous oxide and methane, 1998 - ongoing

The Baltimore Ecosystem Study (BES) established a network of long-term permanent biogeochemical study plots in 1998. These plots provide long-term data on vegetation, soil and hydrologic processes in the key ecosystem types within the urban ecosystem. The network of study plots includes forest plots (upland and riparian), chosen to represent the range of forest conditions in the area and grass plots (to represent home lawns). Plots are instrumented with lysimeters (drainage and tension) to sample soil solution chemistry, time domain reflectometry probes to measure soil moisture, dataloggers to measure and record soil temperature, and trace gas flux chambers to measure the flux of carbon dioxide, nitrous oxide and methane from soil to the atmosphere. Measurements of in situ nitrogen mineralization, nitrification and denitrification were made at approximately monthly intervals from Fall 1998 - Fall 2000. Detailed vegetation characterization (all layers) was done in summer 1998 and 2015. Data from these plots has been published in Groffman et al. (2006, 2009), Groffman and Pouyat (2009), Savva et al. (2010), Costa and Groffman (2013), Duncan et al. (2013), Waters et al. (2014), Ni and Groffman (2018), Templeton et al. (2019). Literature Cited Costa, K.H. and P.M. Groffman. 2013. Factors regulating net methane flux in urban forests and grasslands. Soil Science Society of America Journal 77:850 - 855. Duncan, J. M., L. E. Band, and P. M. Groffman. 2013. Towards closing the watershed nitrogen budget: Spatial and temporal scaling of denitrification. Journal of Geophysical Research Biogeosciences 118:1-5; DOI: 10.1002/jgrg.20090 Groffman PM, Pouyat RV, Cadenasso ML, Zipperer WC, Szlavecz K, Yesilonis IC,. Band LE and Brush GS. 2006. Land use context and natural soil controls on plant community composition and soil nitrogen and carbon dynamics in urban and rural forests. Forest Ecology and Management 236:177-192. Groffman, P.M., C.O. Williams, R.V. Pouyat, L.E. Band and I.C. Yesilonis. 2009. Nitrate leaching and nitrous oxide flux in urban forests and grasslands. Journal of Environmental Quality 38:1848-1860. Groffman, P.M. and R.V. Pouyat. 2009. Methane uptake in urban forests and lawns. Environmental Science and Technology 43:5229-5235. DOI: 10.1021/es803720h. Ni, X. and P.M. Groffman. 2018. Declines in methane uptake in forest soils. Proceedings of the National Academies of Science of the United States of America 115:8587-8590. Savva, Y., K. Szlavecz, R. V. Pouyat, P. M. Groffman, and G. Heisler. 2010. Effects of land use and vegetation cover on soil temperature in an urban ecosystem. Soil Science Society of America Journal 74:469-480. Templeton, L., M.L. Cadenasso, J. Sullivan, M. Neel and P.M. Groffman. 2019. Changes in vegetation structure and composition of urban and rural forest patches in Baltimore from 1998 to 2015. Forest Ecology and Management. In press. Waters, E.R., J.L. Morse, N.D. Bettez and P.M. Groffman. 2014. Differential carbon and nitrogen controls of denitrification in riparian zones and streams along an urban to exurban gradient. Journal of Environmental Quality 43:955–963.

Groffman, Peter M↗
DOE OSTI2021

Volatile aldehyde emissions from “sub-ohm” vaping devices

"Sub-ohm" atomizers with reduced resistance can deliver more power than conventional electronic cigarettes. Typical battery outputs are 100 watts or more. These devices are particularly popular among young users, and can be a significant source of volatile carbonyls in the indoor environment. Emissions from next-generation sub-ohm vaping products were characterized by determining e-liquid consumption and volatile aldehydes emissions for several combinations of popular high-power configurations. Tests explored the effect of dilution air flow (air vent opening), puffing volume, and coil assembly configuration. The mass of liquid consumed per puff increased as the puff volume increased from 50 to 100 mL, then remained relatively constant for larger puff volumes up to 500 mL. This is likely due to mass transfer limitations at the wick and coil assembly, which reduced the vaporization rate at higher puff volumes. Carbonyl emission rates were systematically evaluated using a 0.15 Ω dual coil atomizer as a function of the puffing volume and dilution air flow, adjusted by setting the air vents to either 100% (fully open), 50%, 25%, or 0% (closed). The highest formaldehyde emissions were observed for the lowest puff volume (50 mL) when the vents 31 were closed (48 ng mg-1), opened at 25% (39 ng mg -1 ) and at 50% (32 ng mg -1 ).. By contrast, 50-mL puffs with 100% open vents, and puff volumes >100 mL for any vent aperture, generated formaldehyde yields of 20 ng mg -1 or lower, suggesting that a significant cooling effect resulted in limited carbonyl formation. Considering the effect of the coil resistance when operated at a voltage of 3.8 V, the amount of liquid evaporated per puff decreased as the resistance increased, in the order of 0.15 Ω > 0.25 Ω > 0.6 Ω consistent with decreasing aerosol temperatures measured at the mouthpiece. Three different configurations of 0.15 Ω coils (dual, quadruple and octuple) were evaluated, observing significant variability. No clear trend was found between carbonyl emission rates and coil resistance or configuration, with highest emissions corresponding to a 0.25 Ω dual coil atomizer. Carbonyl emission rates were compared with those determined using the same methodology for conventional e-cigarettes (lower power tank systems), observing overall lower yields for the sub-ohm devices.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

A multifidelity deep operator network approach to closure for multiscale systems

Projection-based reduced order models (PROMs) have shown promise in representing the behavior of multiscale systems using a small set of generalized (or latent) variables. Despite their success, PROMs can be susceptible to inaccuracies, even instabilities, due to the improper accounting of the interaction between the resolved and unresolved scales of the multiscale system (known as the closure problem). In the current work, we interpret closure as a multifidelity problem and use a multifidelity deep operator network (DeepONet) framework to address it. In addition, to enhance the stability and/or accuracy of the multifidelity-based closure, we employ the recently developed ``in-the-loop" training approach from the literature on coupling physics and machine learning models. The resulting approach is tested on shock advection for the one-dimensional viscous Burgers equation and vortex merging for the two-dimensional Navier-Stokes equations. Furthermore, the numerical experiments show significant improvement of the predictive ability of the closure-corrected PROM over the un-corrected one both in the interpolative and the extrapolative regimes.

42 ENGINEERING↗
DOE OSTI2023

Predicting Kyasanur forest disease in resource-limited settings using event-based surveillance and transfer learning

In recent years, the reports of Kyasanur forest disease (KFD) breaking endemic barriers by spreading to new regions and crossing state boundaries is alarming. Effective disease surveillance and reporting systems are lacking for this emerging zoonosis, hence hindering control and prevention efforts. We compared time-series models using weather data with and without Event-Based Surveillance (EBS) information, i.e., news media reports and internet search trends, to predict monthly KFD cases in humans. We fitted Extreme Gradient Boosting (XGB) and Long Short-Term Memory models at the national and regional levels. We utilized the rich epidemiological data from endemic regions by applying Transfer Learning (TL) techniques to predict KFD cases in new outbreak regions where disease surveillance information was scarce. Overall, the inclusion of EBS data, in addition to the weather data, substantially increased the prediction performance across all models. The XGB method produced the best predictions at the national and regional levels. The TL techniques outperformed baseline models in predicting KFD in new outbreak regions. Novel sources of data and advanced machine-learning approaches, e.g., EBS and TL, show great potential towards increasing disease prediction capabilities in data-scarce scenarios and/or resource-limited settings, for better-informed decisions in the face of emerging zoonotic threats.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

Rapid recovery of At-211 by extraction chromatography

The near quantitative separation, purification, and recovery of 211 At has been demonstrated through an extraction chromatography process which utilizes porous beads impregnated with 3-octanone, resulting in an elution yield of 92–95%. Moreover, the rapid nature of this process, <20 min, was achieved after dissolution of a Bi metal target in HNO 3 following retrieval from the beamline after α-particle bombardment. The solution was directly loaded onto the column with no volume or acidity adjustment. The column was washed with HNO 3 and H 2 O, and 211 At was eluted with ethanol, collecting roughly 87–93% in 1 mL. This process of recovering high purity 211 At, in near quantitative yields, represents a significant advance in At separations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Elemental profiling and genomewide association studies reveal genomic variants modulating ionomic composition in Populus trichocarpa leaves

Samples were collected from a population of 1,089 black cottonwood genotypes (P. trichocarpa) assembled from native stands to encompass the central portion of the natural range of the species, stretching from 38.8° to 54.3° N Q13 latitude from California, USA, to British Columbia, Canada. Establishment of the common garden, growth conditions, and site maintenance have been described by Muchero et al (2015). In this study, leaf samples for ionomic profiling were collected from 4-year-old trees, during the growing season, in July 2012, from a field located in Clatskanie, Oregon, USA (46°6′11″N 123°12′13″W). The field site was located in a protected alluvial floodplain containing a uniform Wauna-Locoda silt loam soil area characterized by an acidic pH, in Columbia County, Oregon. A subset of 584 out of the 1,089 P. trichocarpa genotypes were represented in this sampling. These genotypes were randomly selected to represent the geographical distribution of the population. A single fully mature (LPI 7-9) leaf on the south side of the tree exposed to full sunlight conditions was removed from the tree within a 6-hour window centering on solar noon and immediately frozen under dry ice before processing. Leaf samples of 584 P. trichocarpa genotypes were finely ground to 40 mm particle size using a mortar and pestle, and ionomic composition was analyzed using ICP-MS. In total, 20 elements were profiled, including aluminum (Al27), arsenic (As75), boron (B11), cadmium (Cd111), calcium (Ca43), cobalt (Co), copper (Cu), iron (Fe57), magnesium (Mg25), manganese (Mn55), molybdenum (Mo), nickel (Ni60), phosphorus (P31), potassium (K39), rubidium (Rb85), selenium (Se82), sodium (Na23), strontium (Sr88), sulfur (S34), and zinc (Zn66), following a protocol established by Ziegler et al. (2013). For each sample, 75mg of powder was digested overnight in 2.5 mL HNO3 containing 20 parts per billion (ppb) indium as an internal standard, following the protocol described in Ziegler et al. (2013). Following a dilution, concentration of the 20 elements was measured using an Elan 6000 DRC-e mass spectrometer (Perkin-Elmer SCIEX) connected to a PFA microflow nebulizer (Elemental Scientific) and Apex HF desolvator (Elemental Scientific). One measurement per sample per genotype was done. For subsequent analyses, the quantifications were converted to total element concentration.

CBI ionomics, GWAS, plasma-mass spectrometry, neut↗
NASA NTRS2016

BRIC-100VC Biological Research in Canisters (BRIC)-100VC

The Biological Research in Canisters (BRIC) is an anodized-aluminum cylinder used to provide passive stowage for investigations of the effects of space flight on small specimens. The BRIC 100 mm petri dish vacuum containment unit (BRIC-100VC) has supported Dugesia japonica (flatworm) within spring under normal atmospheric conditions for 29 days in space and Hemerocallis lilioasphodelus L. (daylily) somatic embryo development within a 5% CO2 gaseous environment for 4.5 months in space. BRIC-100VC is a completely sealed, anodized-aluminum cylinder (Fig. 1) providing containment and structural support of the experimental specimens. The top and bottom lids of the canister include rapid disconnect valves for filling the canister with selected gases. These specialized valves allow for specific atmospheric containment within the canister, providing a gaseous environment defined by the investigator. Additionally, the top lid has been designed with a toggle latch and O-ring assembly allowing for prompt sealing and removal of the lid. The outside dimensions of the BRIC-100VC canisters are 16.0 cm (height) x 11.4 cm (outside diameter). The lower portion of the canister has been equipped with sufficient storage space for passive temperature and relative humidity data loggers. The BRIC- 100VC canister has been optimized to accommodate standard 100 mm laboratory petri dishes or 50 mL conical tubes. Depending on storage orientation, up to 6 or 9 canisters have been flown within an International Space Station (ISS) stowage locker.

Biological↗
NASA NTRS2016

Kennedy Space Center Fixation Tube (KFT)

Experiments performed on the International Space Station (ISS) frequently require the experimental organisms to be preserved until they can be returned to earth for analysis in the appropriate laboratory facility. The Kennedy Fixation Tube (KFT) was developed to allow astronauts to apply fixative, chemical compounds that are often toxic, to biological samples without the use of a glovebox while maintaining three levels of containment (Fig. 1). KFTs have been used over 200 times on-orbit with no leaks of chemical fixative. The KFT is composed of the following elements: a polycarbonate main tube where the fixative is loaded preflight, the sample tube where the plant or other biological specimens is placed during operations, the expansion plug, actuator, and base plug that provides fixative containment (Fig. 2). The main tube is pre-filled with 25 mL of fixative solution prior to flight. When actuated, the specimen contained within the sample tube is immersed with approximately 22 mL (+/- 2 mL) of the fixative solution. The KFT has been demonstrated to maintain its containment at ambient temperatures, 4degC refrigeration and -100 C freezing conditions.

Space Center↗
NASA NTRS1989

Compact Apparatus Grows Protein Crystals

Laboratory apparatus provides delicately balanced combination of materials and chemical conditions for growth of protein crystals. Apparatus and technique for growth based on hanging-drop method for crystallization of macromolecules. Includes pair of syringes with ganged plungers. One syringe contains protein solution; other contains precipitating-agent solution. Syringes intrude into cavity lined with porous reservoir material saturated with 1 mL or more of similar precipitating-agent solution. Prior to activation, ends of syringes plugged to prevent transport of water vapor among three solutions.

Bugg, Charles E.↗
DOE OSTI2024

Enhancing Docking Accuracy with PECAN2, a 3D Atomic Neural Network Trained without Co-Complex Crystal Structures

Decades of drug development research have explored a vast chemical space for highly active compounds. The exponential growth of virtual libraries enables easy access to billions of synthesizable molecules. Computational modeling, particularly molecular docking, utilizes physics-based calculations to prioritize molecules for synthesis and testing. Nevertheless, the molecular docking process often yields docking poses with favorable scores that prove to be inaccurate with experimental testing. To address these issues, several approaches using machine learning (ML) have been proposed to filter incorrect poses based on the crystal structures. However, most of the methods are limited by the availability of structure data. Here, we propose a new pose classification approach, PECAN2 (Pose Classification with 3D Atomic Network 2), without the need for crystal structures, based on a 3D atomic neural network with Point Cloud Network (PCN). The new approach uses the correlation between docking scores and experimental data to assign labels, instead of relying on the crystal structures. We validate the proposed classifier on multiple datasets including human mu, delta, and kappa opioid receptors and SARS-CoV-2 Mpro. Our results demonstrate that leveraging the correlation between docking scores and experimental data alone enhances molecular docking performance by filtering out false positives and false negatives.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS2011

Biomechanical Analysis of Treadmill Locomotion on the International Space Station

Treadmill locomotion exercise is an important aspect of ISS exercise countermeasures. It is widely believed that an optimized treadmill exercise protocol could offer benefits to cardiovascular and bone health. If training heart rate is high enough, treadmill exercise is expected to lead to improvements in aerobic fitness. If impact or bone loading forces are high enough, treadmill exercise may be expected to contribute to improved bone outcomes. Ground-based research suggests that joint loads increase with increased running speed. However, it is unknown if increases in locomotion speed results in similar increases in joint loads in microgravity. Although data exist regarding the biomechanics of running and walking in microgravity, a majority were collected during parabolic flight or during investigations utilizing a microgravity analog. The Second Generation Treadmill (T2) has been in use on the International Space Station (ISS) and records the ground reaction forces (GRF) produced by crewmembers during exercise. Biomechanical analyses will aid in understanding potential differences in typical gait motion and allow for modeling of the human body to determine joint and muscle forces during exercise. By understanding these mechanisms, more appropriate exercise prescriptions can be developed that address deficiencies. The objective of this evaluation is to collect biomechanical data from crewmembers during treadmill exercise prior to and during flight. The goal is to determine if locomotive biomechanics differ between normal and microgravity environments and to determine how combinations of subject load and speed influence joint loading during in-flight treadmill exercise. Further, the data will be used to characterize any differences in specific bone and muscle loading during locomotion in these two gravitational conditions. This project maps to the HRP Integrated Research Plan risks including Risk of Bone Fracture (Gap B15), Risk of Early Onset Osteoporosis Due to Spaceflight (Gap B15), Risk of Impaired Performance Due to Reduced Muscle Mass, Strength, and Endurance (Gaps M3, M4, M6, Ml, M8, M9) and Risk of reduced Physical Performance Capabilities Due to Reduce Aerobic Capacity (Gaps M7, M8, M9).

De Witt, J. K.↗
DOE OSTI2024

Models, data, and scripts associated with “Prediction of Distributed River Sediment Respiration Rates using Community-Generated Data and Machine Learning”

This data package is associated with the publication “Prediction of Distributed River Sediment Respiration Rates using Community-Generated Data and Machine Learning’’ submitted to the Journal of Geophysical Research: Machine Learning and Computation (Scheibe et al. 2024). River sediment respiration observations are expensive and labor intensive to obtain and there is no physical model for predicting this quantity. The Worldwide Hydrobiogeochemisty Observation Network for Dynamic River Systems (WHONDRS) observational data set (Goldman et al.; 2020) is used to train machine learning (ML) models to predict respiration rates at unsampled sites. This repository archives training data, ML models, predictions, and model evaluation results for the purposes of reproducibility of the results in the associated manuscript and community reuse of the ML models trained in this project. One of the key challenges in this work was to find an optimum configuration for machine learning models to work with this feature-rich (i.e. 100+ possible input variables) data set. Here, we used a two-tiered approach to managing the analysis of this complex data set: 1) a stacked ensemble of ML models that can automatically optimize hyperparameters to accelerate the process of model selection and tuning and 2) feature permutation importance to iteratively select the most important features (i.e. inputs) to the ML models. The major elements of this ML workflow are modular, portable, open, and cloud-based, thus making this implementation a potential template for other applications. This data package is associated with the GitHub repository found at Please see the file level metadata (flmd; “sl-archive-whondrs_flmd.csv”) for a list of all files contained in this data package and descriptions for each. Please see the data dictionary (dd; “sl-archive-whondrs_dd.csv”) for a list of all column headers contained within comma separated value (csv) files in this data package and descriptions for each. The GitHub repository is organized into five top-level directories: (1) “input_data” holds the training data for the ML models; (2) “ml_models” holds machine learning models trained on the data in “input_data”; (3) “scripts” contains data preprocessing and postprocessing scripts and intermediate results specific to this data set that bookend the ML workflow; (4) “examples” contains the visualization of the results in this repository including plotting scripts for the manuscript (e.g., model evaluation, FPI results) and scripts for running predictions with the ML models (i.e., reusing the trained ML models); (5) “output_data” holds the overall results of the ML model on that branch. Each trained ML model resides on its own branch in the repository; this means that inputs and outputs can be different branch-to-branch. Furthermore, depending on the number of features used to train the ML models, the preprocessing and postprocessing scripts, and their intermediate results, can also be different branch-to-branch. The “main-*” branches are meant to be starting points (i.e. trunks) for each model branch (i.e. sprouts). Please see the Branch Navigation section in the top-level README.md in the GitHub repository for more details. There is also one hidden directory “.github/workflows”. This hidden directory contains information for how to run the ML workflow as an end-to-end automated GitHub Action but it is not needed for reusing the ML models archived here. Please the top-level README.md in the GitHub repository for more details on the automation.

13C↗
DOE OSTI2021

Subsurface Characterization and Machine Learning Predictions at Brady Hot Springs Results

Geothermal power plants typically show decreasing heat and power production rates over time. Mitigation strategies include optimizing the management of existing wells - increasing or decreasing the fluid flow rates across the wells - and drilling new wells at appropriate locations. The latter is expensive, time-consuming, and subject to many engineering constraints, but the former is a viable mechanism for periodic adjustment of the available fluid allocations. Data and supporting literature from a study describing a new approach combining reservoir modeling and machine learning to produce models that enable strategies for the mitigation of decreased heat and power production rates over time for geothermal power plants. The computational approach used enables translation of sets of potential flow rates for the active wells into reservoir-wide estimates of produced energy and discovery of optimal flow allocations among the studied sets. In our computational experiments, we utilize collections of simulations for a specific reservoir (which capture subsurface characterization and realize history matching) along with machine learning models that predict temperature and pressure timeseries for production wells. We evaluate this approach using an "open-source" reservoir we have constructed that captures many of the characteristics of Brady Hot Springs, a commercially operational geothermal field in Nevada, USA. Selected results from a reservoir model of Brady Hot Springs itself are presented to show successful application to an existing system. In both cases, energy predictions prove to be highly accurate: all observed prediction errors do not exceed 3.68% for temperatures and 4.75% for pressures. In a cumulative energy estimation, we observe prediction errors that are less than 4.04%. A typical reservoir simulation for Brady Hot Springs completes in approximately 4 hours, whereas our machine learning models yield accurate 20-year predictions for temperatures, pressures, and produced energy in 0.9 seconds. This paper aims to demonstrate how the models and techniques from our study can be applied to achieve rapid exploration of controlled parameters and optimization of other geothermal reservoirs. Includes a synthetic, yet realistic, model of a geothermal reservoir, referred to as open-source reservoir (OSR). OSR is a 10-well (4 injection wells and 6 production wells) system that resembles Brady Hot Springs (a commercially operational geothermal field in Nevada, USA) at a high level but has a number of sufficiently modified characteristics (which renders any possible similarity between specific characteristics like temperatures and pressures as purely random). We study OSR through CMG simulations with a wide range of flow allocation scenarios. Includes a dataset with 101 simulated scenarios that cover the period of time between 2020 and 2040 and a link to the published paper about this project, where we focus on the Machine Learning work for predicting OSR's energy production based on the simulation data, as well as a link to the GitHub repository where we have published the code we have developed (please refer to the repository's readme file to see instructions on how to run the code). Additional links are included to associated work led by the USGS to identify geologic factors associated with well productivity in geothermal fields. Below are the high-level steps for applying the same modeling + ML process to other geothermal reservoirs: 1. Develop a geologic model of the geothermal field. The location of faults, upflow zones, aquifers, etc. need to be accounted for as accurately as possible 2. The geologic model needs to be converted to a reservoir model that can be used in a reservoir simulator, such as, for instance, CMG STARS, TETRAD, or FALCON 3. Using native state modeling, the initial temperature and pressure distributions are evaluated, and they become the initial conditions for dynamic reservoir simulations 4....

15 GEOTHERMAL ENERGY↗