Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “active power”

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 1,351 records · Page 75
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

TCR Postulated Accident and MHA Dose Assessment

Oak Ridge National Laboratory (ORNL) is engaged in regulatory activities to gain US Department of Energy (DOE) approval to build and operate a small high-temperature gas reactor, the Transformational Challenge Reactor (TCR). ORNL uses advanced manufacturing techniques to produce select reactor components. The TCR is unique in its mission: the reactor’s at-power operating lifetime will be less than one effective full-power day to limit the radiological source term to very low values, thus facilitating approval to operate. The primary regulatory documents governing construction and operation of the facility are NUREG-1537, “Guidelines for Preparing and Reviewing Applications for the Licensing of Non-Power Reactors” and DOE-STD-3009, “Preparation of Nonreactor Nuclear Facility Documented Safety Analysis.” Although this document is not a safety basis document, it provides a conservative indication of the magnitude of onsite and offsite radiological doses for specific scenarios, thus informing future design and regulatory activities. As required by NUREG-1537, scenarios included in this analysis are (1) a worstcase depressurized loss-of-forced-circulation accident, and (2) a maximum hypothetical accident. Details on the analysis of dose consequences with various system configurations are also provided.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2026

System Study: Auxiliary Feedwater 1998-2024

This report presents an unreliability evaluation of the auxiliary feedwater (AFW) system at 62 U.S. commercial operating nuclear reactors. New Standardized Plant Analysis Risk (SPAR) models with the most recent SPAR parameter update results were used in this report. Demand, run hour, and failure data from calendar years 1998 to 2024 for selected components were obtained from the Institute of Nuclear Power Operations Industry Reporting and Information System. The unreliability results are trended for the most recent 10 year period and yearly estimates for system unreliability are provided for the entire active period. No statistically significant increasing or decreasing trends were identified in the industry-wide estimates of AFW system start-only unreliability, but a statistically significant decreasing trend was identified in the industry-wide estimates of AFW system 24-hour mission unreliability.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Stellar Activity Manifesting at a One-year Alias Explains Barnard b as a False Positive

Barnard’s star is among the most studied stars given its proximity to the Sun. It is often considered the radial velocity (RV) standard for fully convective stars due to its RV stability and equatorial decl. Recently, an Msin i=3.3M{sub ⊕} super-Earth planet candidate with a 233 day orbital period was announced by Ribas et al. New observations from the near-infrared Habitable-zone Planet Finder (HPF) Doppler spectrometer do not show this planetary signal. We ran a suite of experiments on both the original data and a combined original + HPF data set. These experiments include model comparisons, periodogram analyses, and sampling sensitivity, all of which show the signal at the proposed period of 233 days is transitory in nature. The power in the signal is largely contained within 211 RVs that were taken within a 1000 day span of observing. Our preferred model of the system is one that features stellar activity without a planet. We propose that the candidate planetary signal is an alias of the 145 day rotation period. This result highlights the challenge of analyzing long-term, quasi-periodic activity signals over multiyear and multi-instrument observing campaigns.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

Learning Distribution Grid Topologies: A Tutorial

Unveiling feeder topologies from data is of paramount importance to advance situational awareness and proper utilization of smart resources in power distribution grids. This tutorial summarizes, contrasts, and establishes useful links between recent works on topology identification and detection schemes that have been proposed for power distribution grids. The primary focus is to highlight methods that overcome the limited availability of measurement devices in distribution grids, while enhancing topology estimates using conservation laws of power-flow physics and structural properties of feeders. Grid data from phasor measurement units or smart meters can be collected either passively in the traditional way, or actively, upon actuating grid resources and measuring the feeder's voltage response. Analytical claims on feeder identifiability and detectability are reviewed under disparate meter placement scenarios. Such topology learning claims can be attained exactly or approximately so via algorithmic solutions with various levels of computational complexity, ranging from least-squares fits to convex optimization problems, and from polynomial-time searches over graphs to mixed-integer programs. Although the emphasis is on radial single-phase feeders, extensions to meshed and/or multiphase circuits are sometimes possible and discussed. Here this tutorial aspires to provide researchers and engineers with knowledge of the current state-of-the-art in tractable distribution grid learning and insights into future directions of work.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2020

P-block single-metal-site tin/nitrogen-doped carbon fuel cell cathode catalyst for oxygen reduction reaction

This contribution reports the discovery and analysis of a p-block Sn-based catalyst for the electroreduction of molecular oxygen in acidic conditions at fuel cell cathodes; the catalyst is free of platinum-group metals and contains single-metal-atom actives sites coordinated by nitrogen. The prepared SnNC catalysts meet and exceed state-of-the-art FeNC catalysts in terms of intrinsic catalytic turn-over frequency and hydrogen–air fuel cell power density. The SnNC-NH 3 catalysts displayed a 40–50% higher current density than FeNC-NH 3 at cell voltages below 0.7 V. Additional benefits include a highly favourable selectivity for the four-electron reduction pathway and a Fenton-inactive character of Sn. Here, a range of analytical techniques combined with density functional theory calculations indicate that stannic Sn(iv)Nx single-metal sites with moderate oxygen chemisorption properties and low pyridinic N coordination numbers act as catalytically active moieties. The superior proton-exchange membrane fuel cell performance of SnNC cathode catalysts under realistic, hydrogen–air fuel cell conditions, particularly after NH 3 activation treatment, makes them a promising alternative to today’s state-of-the-art Fe-based catalysts.

25 ENERGY STORAGE↗
DOE OSTI2022

A high-performance and durable direct NH 3 tubular protonic ceramic fuel cell integrated with an internal catalyst layer

Nickel-based cermet anode-supported protonic ceramic fuel cells (PCFCs) show great potential for direct utilization of ammonia. However, the insufficient activity of anode and the deterioration of anode activity/durability caused by the undesired interaction between nickel and ammonia greatly limit the application. Here, we report tubular PCFCs embedded with a catalytic iron layer. Such cells show peak power densities of 1.507 W cm -2 and 1.078 W cm -2 at 700 °C when using H 2 and NH 3 as fuel, respectively, which are the highest tubular PCFC performance so far ever reported. In addition, the stability of cells with the catalyst layer has been dramatically enhanced when compared with that of cells without the catalyst layer. As a result, the enhancement of activity and durability is attributed to the catalytic activity of iron for ammonia decomposition, through which the direct contact between nickel and ammonia has been minimized and the anode structure has therefore been protected.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Deep Learning Systems for Increased Safeguards Surveillance Review Productivity

Nuclear safeguards inspectors expend significant time and maintain intense focus in reviewing video surveillance for safeguards relevant events. To increase efficiency and reduce the time burden of safeguards inspectors performing surveillance data review, this paper presents a novel deep learning (DL) systems concept to integrate generalized DL models into the safeguards surveillance review workflow. The Agency is investigating several DL algorithms for object recognition, localization, tracking, and flagging relevant activities. The project team is working closely with nuclear safeguards inspectors to identify review use cases (based on specific safeguards objectives) and collect their associated requirements. We focused on CANDU and LWR Nuclear Power Plants (NPPs) and their associated dry storage areas as these present a particularly heavy burden on the inspector surveillance review process due to the number of these facilities under safeguards worldwide. Initial DL algorithm results on safeguards data are promising. Using a convolutional neural network, the team attained a mean average precision (mAP) of 92.9% identifying and localizing spent fuel (SF) casks from a 475 surveillance image dataset. Further, the team had initial success in training a recurrent neural network to identify reactor area activities in video clips, successfully indicating when SF casks enter or exit a pool. We discuss how such DL algorithms would be integrated into the Next Generation Surveillance Review (NGSR) software application. Another issue impacting review productivity is the long time inspectors may have to wait when running these algorithms in NGSR. We present a concept to pre-process remotely collected surveillance data with DL models as the data arrives to IAEA headquarters so that results are already available when starting a new review in NGSR. The proposed DL system concept shows a pathway and workflow for increasing an inspector’s surveillance review productivity by quickly and accurately identifying declared and undeclared safeguards relevant objects and activities in large quantities of surveillance imagery data.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2022

C−H Activation of Inert Arenes using a Photochemically Activated Guanidinato‐Magnesium(I) Compound

Abstract UV irradiation of solutions of a guanidinate coordinated dimagnesium(I) compound, [{(Priso)Mg} 2 ] 3 (Priso=[(DipN) 2 CNPr i 2 ] − , Dip=2,6‐diisopropylphenyl), in either benzene, toluene, the three isomers of xylene, or mesitylene, leads to facile activation of an aromatic C−H bond of the solvent in all cases, and formation of aryl/hydride bridged magnesium(II) products, [{(Priso)Mg} 2 (μ‐H)(μ‐Ar)] 4 – 9 . In contrast to similar reactions reported for β‐diketiminate coordinated counterparts of 3 , these C−H activations proceed with little regioselectivity, though they are considerably faster. Reaction of 3 with an excess of the pyridine, p ‐NC 5 H 4 Bu t (py Bu t ), gave [(Priso)Mg(py Bu t H)(py Bu t ) 2 ] 10 , presumably via reduction of the pyridine to yield a radical intermediate, [(Priso)Mg(py Bu t ⋅)(py Bu t ) 2 ] 11 , which then abstracts a proton from the reaction solvent or a reactant. DFT calculations suggest two possible pathways to the observed arene C−H activations. One of these involves photochemical cleavage of the Mg−Mg bond of 3 , generating magnesium(I) doublet radicals, (Priso)Mg⋅. These then doubly reduce the arene substrate to give “Birch‐like” products, which subsequently rearrange via C−H activation of the arene. Circumstantial evidence for the photochemical generation of transient magnesium radical species includes the fact that irradiation of a cyclohexane solution of 3 leads to an intramolecular aliphatic C−H activation process and formation of an alkyl‐bridged magnesium(II) species, [{Mg(μ‐Priso −H )} 2 ] 12 . Furthermore, irradiation of a 1 : 1 mixture of 3 and the β‐diketiminato dimagnesium(I) compound, [{( Dip Nacnac)Mg} 2 ] ( Dip Nacnac=[HC(MeCNDip) 2 ] − ), effects a “scrambling” reaction, and the near quantitative formation of an unsymmetrical dimagnesium(I) compound, [(Priso)Mg−Mg( Dip Nacnac)] 13 . Finally, the EPR spectrum (77 K) of a glassed solution of UV irradiated 3 is dominated by a broad featureless signal, indicating the presence of a doublet radical species.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Ultra High-Temperature Magnetic Bearing System for s-CO 2 Turbines/Expanders. Phase II Final Report

The goal of this project is to design and construct of a prototype of ultra-high temperature permanent magnet (UHT–PM), which provides the force to pull the spinning shaft towards the target position in the magnetic bearing (MB) clearance circle. This project will develop a PM biased MB system, that is actively cooled to the 550°C maximum operating temperature limit of PMs. The MB test environment will include 700°C, 40krpm and 4,000 psi requirements to advance the supercritical carbon dioxide (sCO 2 ) and other types of Brayton power cycle systems. The proposed novel MBs combine PMs for supplying the bias field and electromagnets for supplying the control field, both in UHT conditions. The choice of material is the UHT Sm-Co PMs (US patent 6,451,132) due to their significantly better thermal stability. The 6-pole homopolar design permits uninterrupted force delivery via a redundant control capability (US patent 7,429,811) even if 3 poles were to fail. The Phase I efforts was to focus on development and demonstration of component technologies and detailed design of full scale components and a demonstrator test rig for use in Phase II. UHT–PMs were produced and used in the MB actuator. The operating environment of the EEC-T550 Sm-Co magnet material could be extended by implementing active cooling. Electromagnetic coils imbedded in the MB provide forces to stabilize the shaft position at the target via sensing and control stages. Shaft position sensors enable to identify the present position of the spinning shaft relative to its desired target position in the MB clearance circle (air gap). Auxiliary bearings will be installed to provide a backup support system in the event that power to the MB fails or there is a fault in a device providing current to the MB coils. For control, reliability and safety reasons the sensors and auxiliary bearings are typically located as close as possible to the MBs, which exposes them to the same environmental conditions. Development of only the actuator for the UHT environment is an inadequate approach to complete integration of the MB system in the turbine/expander. Therefore the UHT sensor and auxiliary bearing were designed and developed along with the actively cooled, UHT-PM actuator development. Although shaft position sensors are commercially available (Kaman) they may cost up to $10k per channel, and a full 5 axis magnetic suspension is best implemented with up to 20 sensors. Once successfully completed, this project will significantly extend the operating environment capability for turbomachinery applications such as nuclear power, concentrated solar thermal, fossil fuel, geothermal, and shipboard propulsion. The further advancement of UHT-PM-MB will also strengthen the US position in PM technology, which is presently heavily dominated by China. The UHT Sm-Co magnet material that will be utilized in the proposed work has been developed exclusively by the project proposer. Many actuators, motors, generators and other MB applications will greatly benefit from the proposed work.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2020

Workshop Report for Next-Gen AI for Proliferation Detection: Accelerating the Development and Use of Explainability Methods to Design AI Systems Suitable for Nonproliferation Mission Applications

Artificial intelligence (AI) promises powerful new capabilities in an expansive array of applications. One area is proliferation detection, where AI can provide transformative tools to achieve objectives currently inaccessible using conventional methods. In particular, AI affords the opportunity to use new indicators and process massive amounts of heterogeneous data that can increase sensitivity to proliferant activities and push proliferation detection to the earliest possible stages. However, existing AI tools are not ready to achieve such potential, falling short in regard to some important considerations involved in proliferation detection. One crucial aspect that requires improvement is the explainability of the most powerful AI algorithms (i.e., understanding how the algorithms actually arrive at their conclusions)-the lack of which prevents widespread adoption in national security missions. The Data Science and AI portfolio within the National Nuclear Security Administration's Office of Defense Nuclear Nonproliferation Research and Development is driving the development of next-generation AI for proliferation detection through the expertise and work of the national laboratories and partners in academia. The "Next-Gen AI for Proliferation Detection: Accelerating the Development and Use of Explainability Methods to Design AI Systems Suitable for Nonproliferation Mission Applications" workshop aimed to further this objective. The meeting was held virtually on Sept. 15-16, 2020 and included more than 170 participants, primarily from the national laboratories with additional contributions from university researchers and mission partners. Each day included a keynote presentation, four technical presentations about cutting-edge research in explainability, and a panel to explore considerations in applying explainability and developing AI systems that can accelerate the transition of emerging AI technologies among partners and end users to solve critical mission questions. This report summarizes the Next-Gen AI for Proliferation Detection workshop's content and findings.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2020

Optimal Orbital Selection for Full Configuration Interaction (OptOrbFCI): Pursuing the Basis Set Limit under a Budget

Full configuration interaction (FCI) solvers are limited to small basis sets due to their expensive computational costs. An optimal orbital selection for FCI (OptOrbFCI) is proposed to boost the power of existing FCI solvers to pursue the basis set limit under a computational budget. The optimization problem coincides with that of the complete active space SCF method (CASSCF), while OptOrbFCI is algorithmically quite different. OptOrbFCI effectively finds an optimal rotation matrix via solving a constrained optimization problem directly to compress the orbitals of large basis sets to one with a manageable size, conducts FCI calculations only on rotated orbital sets, and produces a variational ground-state energy and its wave function. Coupled with coordinate descent full configuration interaction (CDFCI), we demonstrate the efficiency and accuracy of the method on the carbon dimer and nitrogen dimer under basis sets up to cc-pV5Z. We also benchmark the binding curve of the nitrogen dimer under the cc-pVQZ basis set with 28 selected orbitals, which provide consistently lower ground-state energies than the FCI results under the cc-pVDZ basis set. Furthermore, the dissociation energy in this case is found to be of higher accuracy.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Aerosol-boundary-layer-monsoon interactions amplify semi-direct effect of biomass smoke on low cloud formation in Southeast Asia

Low clouds play a key role in the Earth-atmosphere energy balance and influence agricultural production and solar-power generation. Smoke aloft has been found to enhance marine stratocumulus through aerosol-cloud interactions, but its role in regions with strong human activities and complex monsoon circulation remains unclear. Here we show that biomass burning aerosols aloft strongly increase the low cloud coverage over both land and ocean in subtropical southeastern Asia. The degree of this enhancement and its spatial extent are comparable to that in the Southeast Atlantic, even though the total biomass burning emissions in Southeast Asia are only one-fifth of those in Southern Africa. We find that a synergetic effect of aerosol-cloud-boundary layer interaction with the monsoon is the main reason for the strong semi-direct effect and enhanced low cloud formation in southeastern Asia.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

High-performance ionomerless cathode anion-exchange membrane fuel cells with ultra-low-loading Ag–Pd alloy electrocatalysts

Rapid translation of catalysts from fundamental studies to high-performance devices could facilitate the process of developing and commercializing anion-exchange membrane fuel cells (AEMFCs). Traditionally, translation from material screening in three-electrode rotating disk electrode (RDE) cells to AEMFCs is complicated by differences in microenvironments, e.g. solid ionomer/membrane vs liquid electrolyte. Herein, we introduce a platform for translation to devices that utilizes ionomerless ultra-low-loading Ag-Pd alloy electrocatalyst cathodes synthesized by co-physical vapor deposition (PVD). Our ionomerless cathodes allow for systematic H 2 -O 2 AEMFC experiments while demonstrating comparable activity trends to those in three-electrode cells. Furthermore, here we show that our Ag 10 Pd 90 -based AEMFC reaches a peak power density of ~1 W cm$_{geo}^{-2}$ and ~10 W mg$_{PGM Cathode}^{-1}$ satisfying the U.S. Department of Energy’s platinum-group-metal (PGM) loading and cost targets. Our approach shows promise in facilitating the rapid translation between three-electrode studies and AEMFCs, offering a simple and effective design for decreasing PGM loadings.

25 ENERGY STORAGE↗
DOE OSTI2021

Strain and ligand effects in Pt-Ni alloys studied by valence-to-core X-ray emission spectroscopy

Experimental detection of the Pt 5d densities of states in the valence band is conducted on a series of Pt-Ni alloys by high energy resolution valence-to-core X-ray emission spectroscopy (VTC-XES) at the Pt L 3 -edge. VTC-XES measurements reveal that the Pt d-band centroid shifts away from the Fermi level upon dilution, accompanied by concentration-dependent Pt d-band width. The competition between the strain effect and ligand effect is observed experimentally for the first time. It is found that the d-band widths in Pt 3 Ni and PtNi are broader than that of Pt metal due to compressive strain which overcompensates the effect of dilution, while it is narrower in PtNi 3 where the ligand effect dominates. VTC-XES is demonstrated to be a powerful tool to study the Pt d-band contribution to the valence band of Pt-based bimetallic. The implication for the enhanced activity of Pt-Ni catalysts in oxygen reduction reaction is discussed.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Unified understanding of the impact of semiflexibility, concentration, and molecular weight on macromolecular-scale ring diffusion

Conformationally fluctuating, globally compact macromolecules such as polymeric rings, single-chain nanoparticles, microgels, and many-arm stars display complex dynamic behaviors due to their rich topological structure and intermolecular organization. Synthetic rings are hybrid objects with conformations that display both ideal random walk and compact globular features, which can serve as models of genomic DNA. To date, emphasis has been placed on the effect of ring molecular weight on their unusual behaviors. Here, we combine simulations and a microscopic force-level theory to build a unified understanding for how key aspects of ring dynamics depend on different tunable molecular properties including backbone rigidity, monomer concentration, degree of traditional entanglement, and molecular weight. Our large-scale molecular dynamics simulations of ring melts with very different backbone stiffnesses reveal unanticipated behaviors which agree well with our generalized theory. This includes a universal master curve for center-of-mass diffusion constants as a function of molecular weight scaled by a chemistry and thermodynamic state-dependent critical molecular weight that generalizes the concept of an entanglement cross-over for linear chains. The key physics is how backbone rigidity and monomer concentration induced changes of the entanglement length, interring packing, degree of interpenetration, and liquid compressibility slow down space-time dynamic-force correlations on macromolecular scales. A power law decay of the center-of-mass diffusion constant with inverse molecular weight squared is the first consequence, followed by an ultraslow activated hopping transport regime. Our results set the stage to address slow dynamics and kinetic arrest in different families of compact synthetic and biological polymeric systems.

Science & Technology - Other Topics↗
DOE OSTI2026

Expert evaluation of LLM world models: A high-T c superconductivity case study

Large Language Models (LLMs) show great promise as a powerful tool for scientific literature exploration. However, their effectiveness in providing scientifically accurate and comprehensive answers to complex questions within specialized domains remains an active area of research. Using the field of high-temperature cuprates as an exemplar, we evaluate the ability of LLM systems to understand the literature at the level of an expert. We construct an expert-curated database of 1,726 scientific papers that covers the history of the field, and a set of 67 expert-formulated questions that probe deep understanding of the literature. We then evaluate six different LLM-based systems for answering these questions, including both commercially available closed models and a custom retrieval-augmented generation (RAG) system capable of retrieving images alongside text. Experts then evaluate the answers of these systems against a rubric that assesses balanced perspectives, factual comprehensiveness, succinctness, and evidentiary support. Among the six systems, two using RAG on curated literature outperformed existing closed models across key metrics, particularly in providing comprehensive and well-supported answers. We discuss promising aspects of LLM performances as well as critical short-comings of all the models. The set of expert-formulated questions and the rubric will be valuable for assessing expert level performance of LLM based reasoning systems.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Feasibility of raised inner strike point equilibria scenario in ITER for detritiation from beryllium co-deposits

Abstract In ITER, tritium retention primarily occurs through co-deposition with beryllium. To avoid exceeding the strict tritium inventory limit, efficient tritium recovery techniques are essential. Baking is the ITER baseline for tritium recovery, but its effectiveness in removing tritium from thick beryllium layers is limited. A raised strike point scenario is considered an alternative method for removing tritium from the ITER inner vertical divertor target by heating components via plasma flux. This paper presents SOLPS-ITER code simulations conducted under various conditions, assessing the divertor performance and tritium outgassing of the raised strike point scenario. As the strike point is raised, recycled neutrals are not efficiently baffled by the dome and scrape-off layer, significantly changing the neutral trajectory and ionization source distribution. This improves detachment accessibility but worsens core-edge compatibility compared to the baseline scenario. However, in the partially detached condition, the impact of raising the strike point, perpendicular transport, and q 95 on target heat flux is not significant, as it primarily scales with the input power. Target heat flux is translated to target surface temperature using a simplified heat transfer model that considers the 3D target monoblock geometry and active cooling condition, excluding Be layer thermal properties. For partially detached divertor conditions, the bulk tungsten monoblock surface temperature remains below the baking temperature, which is insufficient for efficient tritium outgassing under the actively cooled ITER divertor condition. However, considering the potential thermal contact resistance between the beryllium and tungsten layers, which may significantly impact temperature distribution, the temperature of the beryllium layer can be raised to a level sufficient for efficient tritium outgassing. Therefore, the raised strike point scenario can be considered as an alternative in-vessel tritium removal technique.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Machine learning materials properties with accurate predictions, uncertainty estimates, domain guidance, and persistent online accessibility

One compelling vision of the future of materials discovery and design involves the use of machine learning (ML) models to predict materials properties and then rapidly find materials tailored for specific applications. However, realizing this vision requires both providing detailed uncertainty quantification (model prediction errors and domain of applicability) and making models readily usable. At present, it is common practice in the community to assess ML model performance only in terms of prediction accuracy (e.g. mean absolute error), while neglecting detailed uncertainty quantification and robust model accessibility and usability. Here, we demonstrate a practical method for realizing both uncertainty and accessibility features with a large set of models. We develop random forest ML models for 33 materials properties spanning an array of data sources (computational and experimental) and property types (electrical, mechanical, thermodynamic, etc). All models have calibrated ensemble error bars to quantify prediction uncertainty and domain of applicability guidance enabled by kernel-density-estimate-based feature distance measures. All data and models are publicly hosted on the Garden-AI infrastructure, which provides an easy-to-use, persistent interface for model dissemination that permits models to be invoked with only a few lines of Python code. We demonstrate the power of this approach by using our models to conduct a fully ML-based materials discovery exercise to search for new stable, highly active perovskite oxide catalyst materials.

domain of applicability↗