SEARCH · Search NASA

Results for “Atomistic Materials Modeling”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

713 records · Page 4

All sources NASA NTRS DOE OSTI

NASA NTRSDate not supplied

BEAST: Expanding Sustainable Data Infrastructure for High-Enthalpy Facilities

Reproducible, data-driven thermal protection system (TPS) research requires that experimental records from high-enthalpy testing be consistently structured, traceable, and accessible across campaigns and institutions. In practice, however, arcjet and plasma facilities data remain largely fragmented: raw diagnostics are stored in ad hoc formats, material sample histories are disconnected from test conditions, and metadata standards are absent, precluding systematic cross-campaign analysis and long-term reuse. BEAST (Backend for Experiment Analysis, Storage, and Traceability) is an open-source, web-based platform that addresses these limitations by providing a unified, queryable infrastructure for high-enthalpy ground-test data [1]. First presented at the 15th Ablation Workshop [2], BEAST has since undergone significant development. The platform ingests and structures multi-channel time-series diagnostics, facility configurations, and material property records within a common provenance model, ensuring end-to-end traceability from raw sensor acquisition to reduced experimental quantities. A versioned material library links specimen identity and processing history to the specific runs in which each sample was tested. An integrated modeling workbench enables training and evaluation of regression models directly on archived experimental data, supporting condition interpolation and the construction of empirical material response databases. Beyond its original deployment at NASA Ames Research Center, BEAST has been designed to be facility-agnostic, with ongoing efforts to extend its adoption to other facilities. Its modular architecture accommodates heterogeneous diagnostic setups and facility types, and its future open-source distribution allows institutions to build on a common data standard rather than maintaining isolated, bespoke solutions. BEAST is further integrated within a broader ecosystem of companion tools: arcjetCV [3] extracts recession rates and shock standoff distances from high-speed video using computer vision, and miniSTARscan [4] provides sub-minute, portable photogrammetric surface reconstruction of test articles before and after exposure. All tools share a common data schema, enabling seamless ingestion of surface geometry, imagery, and time-series data into a single, coherent experimental record.

Database↗

NASA NTRS2019

Physics-Based Modeling and Simulation of Emerging Battery Technologies for Aerospace

Recently there is a growing interest in the aviation sector to reduce air and noise pollution. Electrochemical energy storage devices such as batteries coupled with a distributed electric propulsion system can reduce noise concerns as well as emissions and allow the concept of Urban Air Mobility to come to fruition. The battery performance needs to improve considerably compared from current state-of-art Li-ion battery (about 200Wh/Kg) to realize all-electric passenger jet for short flights (up to 690 miles). NASA is exploring lithium-oxygen battery chemistry to power hybrid (battery powered electrical system) and all-electric aircraft for short distance and long-distance flights. Li-O2 is one of the advanced Li-ion technologies that promise to provide specific energy of more than 750Wh/Kg. For this presentation, we present our work on improving power density of Li-O2 batteries through the use of multiphysics simulations. Next, a path is outlined to port these model to simulate performance for a new battery chemistry for space application, Li-CO2. Li-CO2 uses carbon dioxide as the active material instead of oxygen. Although this technology is in its early development, the offers two benefits: it can be used as a CO2 scrubber, as oxygen is one of the by-products on charging, and as a backup or a standalone battery for various Mars or Venus missions, where the carbon dioxide content in the atmosphere is high and need battery to operate at higher temperatures.

Mehta, Mohit↗

NASA NTRS1990

Satellite Material Contaminant Optical Properties

The Air Force Wright Research and Development Center and the Arnold Engineering Development Center are continuing a program for measuring optical effects of satellite material outgassing products on cryo-optic surfaces. Presented here are infrared (4000 to 700 cm -1 ) transmittance data for contaminant films condensed on a 77 K geranium window. From the transmittance data, the contaminant film refractive and absorptive indices (n, k) were derived using an analytical thin-film interference model with a nonlinear least-squares algorithm. To date 19 materials have been studied with the optical contents determined for 13 of those. The materials include adhesives, paints, composites, films, and lubricants. This program is continuing and properties for other materials will be available in the future.

B E Wood↗

NASA NTRSDate not supplied

Design and Analysis of Battery Thermal Management Systems

Thermal management of battery cell packs is a critically needed technology. The purpose of this work is to design new and improved Battery Thermal Management Systems (BTMS) for use in electric airplanes. The BTMS should be 3D printable and hold twelve to sixteen 18650 batteries. To minimize aircraft’s weight, an actively air-cooled battery pack was chosen to avoid the excess weight of water-cooled and phase change material (PCM) cooled packs. As a result, significant changes were made to remove all metal in the pack and replace it with lower density polymer matrix composites (PMCs) which can be additively manufactured. Two air cooled battery pack configurations (traditional propeller fans & bladeless fans) were designed, modeled, and compared. These packs were first modeled in SolidWorks 2021 3D CAD, then imported into COMSOL MultiPhysics to be studied using the “Heat Transfer in Solids and Fluids” module. The design with the bladeless fans eliminated the need to use high conductivity heavy metal to remove unwanted thermal energy. These bladeless fans were designed entirely out of PMCs. This thermal pack design weighs 0.04 kg less than that of the traditional propeller fans design and has increased the battery pack energy density by 8.25 Wh/kg.

Thermal Management System for Battery Packs↗

NASA NTRS2022

Optical properties of morphologically complex black carbon aerosols: Effects of coatings

Optical properties are computed for fractal-like aggregate black carbon (BC) aerosols coated with different substances. Two models are used for these aerosols: (i) the coated aggregate model (model I), where coating is added to the voids and surroundings of BC fractal-like aggregates; and (ii) the closed-cell aggregate model (model II), where coating is added concentrically to each monomer of BC fractal-like aggregates. Our results favor choosing the coated aggregate model (model I) to simulate scattering and absorption by coated BC aerosols because this model is morphologically more realistic, and because this model yields mass absorption cross section (MAC) and backscattering linear depolarization ratio (LDR) values that are consistent with field measurements. Moreover the corresponding computed degree of linear polarization (DoLP) and LDR values are very sensitive to changes in the coating volume fraction (fvol) and the coating refractive index (m). With the same absorbing BC core, the MAC value increases steadily with increasing fvol and m values. For example, using a density of 1.8 g/cm 3 and BC refractive index 1.95 + i0.79, the calculated MAC values for uncoated BC aggregates range from 6.2 to 6.8 m 2 /g at 0.55 µm. When coating material is applied to the BC aggregates, the calculated MAC values for model I particles increase to between 9.8 and 13.2 m 2 /g (depending on m) when fvol = 87.5%. The backscattering LDR values also tend to increase with the increasing m values for the shapes and sizes considered in our study. For model I particles, the backscattering LDR values span a wide range of 4.2–27.8% at a wavelength of 0.35 µm at fvol = 87.5% when m increases from 1.33 to 1.55. Our results are relevant to analyses of polarimetric and lidar observations of smoke particles, especially when these particles undergo hygroscopic growth.

Black carbon aerosols↗

NASA NTRSDate not supplied

From Exploration Flight Test-1 to Artemis II--A NASA Langley's Orion Aerosciences Overview

The Orion Aerosciences program at NASA Langley has played a central role in advancing the aerodynamic and aeroheating prediction capabilities required for the Orion crew vehicle’s return from deep space. This presentation provides a technical overview of aerosciences contributions spanning Exploration Flight Test-1 (EFT-1), Artemis I, and the ongoing post-flight analysis of Artemis II. EFT-1 provided the first high-energy entry dataset for Orion, enabling critical validation of aerodynamic force and moment predictions, static and dynamic stability characteristics, and aeroheating environments at relevant flight Mach and Reynolds numbers. Flight-derived pressure data were used to refine the Flush Air Data System (FADS) methodology for atmospheric density reconstruction and to improve Best Estimated Trajectory (BET) solutions. The EFT-1 data also offered key insights into heat shield performance, including char layer recession, in-depth thermal response, and material retention behavior under flight conditions, informing updates to both thermal response models and uncertainty quantification practices. Building on EFT-1, Artemis I extended the database to true lunar-return conditions. Observations of heat shield performance, including localized char loss, bondline response, and recession variability, provided an unprecedented opportunity to reassess Thermal Protection System (TPS) and aeroheating modeling assumptions. Aerodynamic reconstruction efforts incorporated improved FADS calibration, enhanced atmospheric modeling, and refined force and moment databases to reduce trajectory and load uncertainties. Aeroheating comparisons between pre-flight predictions and flight data enabled targeted model updates, particularly in transitional flow environments and wake heating regions. For Artemis II, these lessons were systematically incorporated into the pre-flight prediction process. Updates included refined aerodynamic databases anchored to flight-validated corrections, improved density estimation and BET methodologies using enhanced database interpolation algorithm and FADS modeling, and revised aeroheating design environments informed by Artemis I material response observations. By the time of the workshop, Artemis II post-flight analysis will be underway, and preliminary findings will be presented where available, including early comparisons of aerodynamic reconstruction, atmospheric density estimation, and thermal protection system performance relative to updated predictions. Collectively, this body of work is a testament to the dedicated and multidisciplinary team whose sustained efforts have contributed to the program’s success and to the progressive maturation of Orion aerosciences modeling through numerical modeling, ground and flight data assimilation. The integrated advancement of aerodynamics, trajectory reconstruction, FADS-based density estimation, and aeroheating analysis has reduced predictive uncertainty and strengthened confidence for future crewed lunar and deep-space missions.

Orion↗

NASA NTRS1977

NASA's Advanced Control Law Program for the F-8 Digital Fly-By-Wire Aircraft

This paper briefly describes the NASA F-8 Digital Fly-By-Wire (DFBW) and Langley Research Center's role in investigating and promoting advanced control laws for possible flight experimentation. The primary purpose of the paper is to provide background material for the papers that follow which describe some of the F-8 DFBW adaptive control law studies. The paper provides a brief description of the Phase II DFBW F-8 aircraft and its control system, discusses some of the advanced control law study objectives and guidelines, and provides some mathematical models which are useful in the control analysis problem.

Aerospace Materials↗

NASA NTRS2026

Development and Validation of a High-Vacuum Thermal Conductivity Testbed for Aerospace Interface Materials

Thermal Interface Materials (TIMs) are critical components in spacecraft thermal management systems, where thermal performance is strongly influenced by vacuum conditions, interface contact resistance, and layered metallic joint behavior. However, manufacturer-reported thermal conductivity values are often derived under idealized conditions and may not accurately represent performance within operational aerospace applications. To address this limitation, the Testbed for Advanced Interface Materials in Vacuum (TAIMV) was developed as a modular vacuum-compatible thermal conductivity characterization platform capable of evaluating aerospace-relevant TIM configurations under both ambient and high-vacuum environments. The testbed was derived from the ASTM C1044-16 guarded hot plate methodology and incorporates interchangeable layers of stainless steel coupon geometries, independently controlled main and guard heaters, embedded resistance temperature detectors (RTDs), thermocouples, multi-layer insulation (MLI), and a temperature-controlled cold plate to characterize through-thickness thermal gradients across layered interfaces. In the current configuration, interface compression is limited to the nominal contact pressure generated by the experimental stack assembly. Initial experimental campaigns were conducted at ambient pressure and below 1×10-5 torr for vacuum cases using multiple interface materials including Braycote 601EF and Krytox-based greases across a range of thermal operating conditions. In parallel, a coupled numerical Python thermal model was developed to predict temperature distribution throughout the stack while accounting for conduction, radiation, and parasitic heat transfer pathways and effective interface resistance effects. Experimental measurements and numerical predictions showed consistent thermal trends across multiple operating conditions and environmental states. Results also revealed measurable differences between ambient and vacuum thermal behavior, demonstrating the importance of interface resistance, parasitic heat transfer mechanisms, and stack geometry in determining effective thermal performance within layered thermal interfaces. The presented work establishes a foundation for future thermal model correlation efforts and expanded characterization of aerospace thermal interface materials under representative environmental conditions. Future work will focus on the integration of a load cell system to enable controlled pressure-dependent characterization of thermal interface materials under compressive loading. This capability will allow investigation of the influence of contact pressure on effective thermal conductivity, interface resistance, and thermal performance within layered aerospace thermal interfaces under representative operational conditions.

Thermal Development Testing↗

NASA NTRS1998

Summary of Research Report

Ten papers, published in various publications, on buckling, and the effects of imperfections on various structures are presented. These papers are: (1) Buckling mode localization in elastic plates due to misplacement in the stiffner location; (2) On vibrational imperfection sensitivity on Augusti's model structure in the vicinity of a non-linear static state; (3) Imperfection sensitivity due to elastic moduli in the Roorda Koiter frame; (4) Buckling mode localization in a multi-span periodic structure with a disorder in a single span; (5) Prediction of natural frequency and buckling load variability due to uncertainty in material properties by convex modeling; (6) Derivation of multi-dimensional ellipsoidal convex model for experimental data; (7) Passive control of buckling deformation via Anderson localization phenomenon; (8)Effect of the thickness and initial im perfection on buckling on composite cylindrical shells: asymptotic analysis and numerical results by BOSOR4 and PANDA2; (9) Worst case estimation of homology design by convex analysis; (10) Buckling of structures with uncertain imperfections - Personal perspective.

Isaac Elishakoff↗

NASA NTRS2004

Ejecta Generation and Redistribution on 433 Eros: Modeling Ejecta Launch Conditions

The NEAR-Shoemaker mission to asteroid 433 Eros presents an unprecedented opportunity to gain fundamental new knowledge about the processes governing regolith formation and redistribution on small bodies. NEAR-Shoemaker’s high-resolution imaging of the surface of Eros makes the asteroid a valuable and heretofore unparalleled laboratory for the detailed study of impact ejecta reaccretion and regolith redistribution on low-gravity (of order 10 -3 g) objects. Regolith is produced on asteroids by impact cratering, and the existence of regolith on the smallest solar system bodies supports the view that some of the ejecta from impact events on such objects may be retained. Impact craters and retained ejecta on low-gravity objects like Eros represent valuable natural laboratories for evaluating various models of impact cratering processes, since they may present crater structures or ejecta features that either do not form or are hidden on higher-gravity bodies like the Moon. Further, quantifying the extent to which impact processes generate and redistribute regoliths on small body surfaces (excavation depths, retained fraction, turnover timescales, etc.) is pivotal to the issue of how to relate meteoritical samples to their asteroidal parent bodies when surficial processes ( i.e., “space weathering”) may disguise or cover up underlying material and confound the ability of remote sensing techniques to provide reliable mineralogical assays of the parent objects. The rich variety of data on Eros’ regolith properties and distribution returned by NEAR-Shoemaker now require detailed analysis in order to take full advantage of the clues these observations offer for elucidating details of the impact cratering process on small bodies. Complicating simple interpretations of crater and ejecta morphology are dynamical effects on ejecta emplacement resulting from Eros’ irregular shape, rapid (5.27 hr) rotation, and low gravity. Figure 1 shows the very different ejecta deposit morphology that can result if the effects of rotation alone are neglected. Considering the additional complicating factors of Eros’ irregular shape and complex gravitational field, simple calculations of the extent and thickness of ejecta blankets and the spatial distribution of ejecta blocks from basic crater scaling laws or numerical hydrocodes alone do not suffice. In order to fully interpret the suite of NEAR-Shoemaker observations of regolith features across the surface of Eros and to evaluate various impact models for specific craters on the asteroid, detailed dynamical modeling of the deposition of crater ejecta from those craters is required . Here, I describe some modifications and improvements to the dynamical model being used for these studies.

D D Durda↗

NASA NTRS2026

Development of a One-Domain Volume-Averaged Navier–Stokes Solver

The interaction between a high-enthalpy flow and a thermal protection material is inherently multiscale and multiphysics. In conventional aerothermal analyses, the external flow and material response are generally modeled using separate computational domains coupled through boundary conditions at the material surface. Although this approach has supported many practical applications, it requires assumptions about the location and behavior of the interface and may become difficult to apply when material decomposition, internal reactions, and surface recession substantially alter the porous structure. This report presents the development of a one-domain formulation in which the free-fluid and porous-material regions are represented within a single computational domain. The formulation is based on the volume-averaged Navier–Stokes (VANS) equations, derived from the governing equations for reacting, compressible flow and condensed material. Volume averaging transfers the influence of the unresolved material microstructure to the macroscale equations through effective transport properties, interfacial source terms, and dispersion fluxes. Particular attention is given to regions in which porosity and permeability vary rapidly, including the diffuse transition between a porous material and the surrounding fluid. The resulting equations are implemented in the Porous-material Analysis Toolbox based on OpenFOAM (PATO). The report describes the pressure–velocity coupling strategy used by the solver, examines spatial filtering techniques for deriving effective properties, and evaluates the influence of a smoothly varying interface permeability. Numerical demonstrations include canonical porous-flow configurations, a flow-tube configuration representative of FiberForm® permeability experiments, and the oxidation of a porous carbon material. The purpose of this work is to establish a mathematical and computational foundation for a unified treatment of flow and thermal protection material response. The present formulation is intended to support the progressive inclusion of additional physical processes, including multicomponent transport, finite-rate gas–surface chemistry, pyrolysis, internal oxidation, and material recession. It also provides a framework for connecting pore-scale simulations and microstructural characterization with macroscale aerothermal-response calculations. This report is intended for researchers and engineers working in computational fluid dynamics, porous-media transport, material response, and thermal protection system modeling. It documents both the theoretical development and the initial numerical assessment of the one-domain approach, while identifying the closure of effective and dispersion terms as an important subject for continued investigation.

Ablation↗

NASA NTRS1995

RTE: A Computer Code for Rocket Thermal Evaluation

The numerical model for a rocket thermal analysis code (RTE) is discussed. RTE is a comprehensive thermal analysis code for thermal analysis of regeneratively cooled rocket engines. The input to the code consists of the composition of fuel/oxidant mixture and flow rates, chamber pressure, coolant temperature and pressure. dimensions of the engine, materials and the number of nodes in different parts of the engine. The code allows for temperature variation in axial, radial and circumferential directions. By implementing an iterative scheme, it provides nodal temperature distribution, rates of heat transfer, hot gas and coolant thermal and transport properties. The fuel/oxidant mixture ratio can be varied along the thrust chamber. This feature allows the user to incorporate a non-equilibrium model or an energy release model for the hot-gas-side. The user has the option of bypassing the hot-gas-side calculations and directly inputting the gas-side fluxes. This feature is used to link RTE to a boundary layer module for the hot-gas-side heat flux calculations.

Mohammad H N Naraghi↗

NASA NTRSDate not supplied

Dust Composition of Comet 81P/Wild 2 From JWST Spectroscopy Compared to Stardust’s Fine-Grained Materials and Gems-Rich IDPS

The Stardust Mission returned samples from the coma of Jupiter Family comet 81P/Wild 2 for detailed laboratory analyses. Here we present and discuss the best-fit thermal dust model for the coma dust of comet 81P/Wild 2 as observed by JWST. Comet 81P was observed through JWST GO 018xx, using NIRSpec (2.9–5.3µm, λ/∆λ≈1000) and MRS IFU (4.9–28.1µm, λ/∆λ≈3000) on 2023-03-20 UT and 2023-03-24 UT, respectively, at a heliocentric distance of 1.85 au and JWST distance of 1.43 au (phase angle of 32 degrees). The dust coma of 81P as revealed by JWST offers a salient compliment to laboratory studies of Stardust samples that typically are bigger than 2 µm and up to 60 µm in size.

D H Wooden↗

NASA NTRS1994

Analysis of the Erosion of Cubic Centered and Face Centered Metallic Materials [Analyse de L'erosion de Matereaux Metalliques de Structures Cubiques Centree et Face Centree]

A generalized law of erosion, in nondimensional form, is proposed. The aforementioned explicitly gives an account of the nature and the microstructure of the material impacted, the angle of impact of the particles, and their hardness. The erosion behavior of various steels, pig iron, and cast iron is examined as well as the behavior of new ordered intermetallic materials, Fe-25% Al-(Zr, Cr, B). The concern of such alloys in an erosion-corrosion environment is shown. --- Une loi généralisée d'usure par érosion, sous forme non dimensionnelle, est proposée. Celle-ci rend compte explicitement de la nature et de la microstructure du matériau impacté, de l'angle d'impact des particules, de leur acuité at de leur durete. Le comportement à l'érosion de divers aciers et fontes est examiné ainsi que la tenue de nouveaux matériaux intermétalliques ordonnés Fe-25% Al - (Zr, Cr, B). L'interet de tels alliages dans des environnements agressifs du type érosion-corrosion est montré. Mots clés: usure, érosion, modèle, matériaux métalliques, intermétalliques Fe-Al.

A Magnée↗

NASA NTRSDate not supplied

Multifaceted Considerations for Fission Surface Power Radiation Shielding Design

The purpose of the Fission Surface Power (FSP) nuclear system is to provide energy on the Moon and Mars for supporting their exploration and colonization. The FSP design and deployment can leverage the past research and development work in areas, such as reactor design, but such crucial components as radiation shielding and instrumentation still need material and manufacturing studies, modeling, and testing. The shielding components are needed for the protection of humans, electronics, and sensitive components. They are essential system components, but they are often the heaviest components of the system for which each kilogram matters. Different material combinations and layouts are possible. The studies presented were focused on the evaluation of several options. The initial analysis included their performance assessment, initial temperature distribution evaluation and manufacturability considerations. Several feasible shield design options were identified and discussed in this paper, as well.

Fission Surface Power↗

NASA NTRS2026

Performance Evaluation of Intravehicular Activity Spacesuit Without the Use of a Liquid Cooling Garment

The Orion Crew Survival Systems (OCSS) suit is equipped with safety technology that protects the crew during launch and re-entry and Intravehicular Activities (IVA). It was designed to be used with a Liquid Cooling Garment (LCG), an undergarment with tubes that circulate water to remove excess heat from a crew member. The objective of this study is to determine if the Modified Advanced Crew Escape Suit (MACES), which has very similar material, assembly, and pressure characteristics as the OCSS suit, can be used without the LCG to maintain the crew within their heat storage requirements where exceedances can lead to cognitive and physiological impairment. Testing was completed for both hot and cold environments where test subjects wore varying configurations of the MACES suit ensemble without the LCG. The tests were conducted in a temperature and humidity-controlled chamber with four test subjects ranging from extra small to large. Target metabolic rates were reached using an arm ergometer. To represent the different suit configurations seen in the concept of operations, the suit was tested with visor-down, visor-up, and without helmet and gloves. The test data was analyzed to determine how the environment, test subject size, metabolic rate, and suit configuration affected heat storage. Furthermore, the test data was used to correlate the METMAN model, a transient metabolic man program that simulates the heat transfer within a crew member’s body and from a crew member to the surrounding environment. The METMAN correlation showed good agreement between predicted heat storage and test data at the end of the test duration, with a coefficient of determination (R 2 ) value of 0.8999 for visor-down cases and R 2 of 0.8902 for visor-up and helmet off cases. A correlated METMAN model allows for the analysis and prediction of IVA suit performance without an LCG in additional scenarios.

Spacesuit↗

NASA NTRS2026

Performance Evaluation of Intravehicular Activity Spacesuit Without the Use of a Liquid Cooling Garment

Human Thermal Modeling↗

NASA NTRS2025

Improving Adhesive Bondline Time of Flight Predictions During Autoclave Cure Utilizing Machine Learning

Composite materials are increasingly being used in aerospace applications due to their superior strength-to-weight ratio compared to commonly used metals. A current limitation to widespread adoption is the certification of adhesively bonded joints. One approach to improving adhesive bonding in composites is accurately measuring the thickness of adhesive bondlines in composite laminates. Precise bondline thickness control is essential for aerospace applications where adhesive layer thickness directly affects joint fracture properties and structural performance. This study focused on implementing machine learning techniques to determine the ultrasonic time of flight (directly correlated to thickness) in adhesive bondlines throughout autoclave cure cycles. A high-temperature (use up to 180°C) ultrasonic scanning system was deployed in an autoclave to provide time of flight data through composite panels. Three experiments were conducted on the curing of 305 mm × 305 mm unidirectional composite panels. In the first experiment, a piecewise function was fit for the temperature correction factor to account for changing autoclave temperatures. Due to deficiencies in the first calibration experiment, a second experiment was run, and the results were used to train a machine learning model. The revised experiment, in combination with the machine learning model, significantly increased the accuracy of the bondline time of flight predictions (~14% error reduced to <1%). Data was processed using the Regression Learner Application in MATLAB®, with a Support Vector Machine selected for the model. The result was a machine learning algorithm capable of reliably quantifying ultrasonic time of flight through adhesive bondlines. The third experiment provided independent test data for the machine learning model, demonstrating that the model produces accurate predictions from data beyond its training set.

Machine Learning↗