SEARCH · Engineering Papers

Results for “system emulation”

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.

At least 451 records · Page 25

All sources NASA NTRS DOE OSTI

NASA NTRS2023

An Examination of Heavy Ion-Induced Persistent Visual Error Signatures in an Electronic Display Driver Integrated Circuit

Given the ubiquity of electronic displays integration in human-based systems, the impending mission critical, space-based applications of electronic displays will necessitate SEE assessment of components unique to electronic displays. A commercially available DDIC designed to drive a small form factor organic light emitting diode (OLED) was visually monitored during heavy ion irradiation to catalogue radiation induced persistent visual error signatures that require manual intervention (i.e., power cycling) to return to nominal function. These error signatures were able to be reproduced via modification of configuration register values utilizing the instruction set intended for interfacing a microcontroller with the DDIC. This approach to emulation of heavy-ion induced errors on a table-top assists with human perception-based criticality analysis as well as development of mitigation techniques.

Electronic Displays↗

NASA NTRS2019

Usability Evaluation of Indicators of Energy-Related Problems in Commercial Airline Flight Decks

A series of pilot-in-the-loop flight simulation studies were conducted at NASA Langley Research Center to evaluate indicators aimed at supporting the flight crew’s awareness of problems related to energy states. Indicators were evaluated utilizing state-of-the-art flight deck systems such as on commercial air transport aircraft. This paper presents results for four technologies: (1) conventional primary flight display speed cues, (2) an enhanced airspeed control indicator, (3) a synthetic vision baseline that provides a flight path vector, speed error, and an acceleration cue, and (4) an aural airspeed alert that triggers when current airspeed deviates beyond a specified threshold from the selected airspeed. Full-mission high-fidelity flight simulation studies were conducted using commercial airline crews. Crews were paired by airline for common crew resource management procedures and protocols. Scenarios spanned a range of complex conditions while emulating several causal factors reported in recent accidents involving loss of energy state awareness by pilots. Data collection included questionnaires administered at the completion of flight scenarios, aircraft state data, audio/video recordings of flight crew, eye tracking, pilot control inputs, and researcher observations. Questionnaire response data included subjective measures of workload, situation awareness, complexity, usability, and acceptability. This paper reports relevant findings derived from subjective measures as well as quantitative measures.

Evans, Emory T.↗

NASA NTRS2012

Integrated RF/Optical Interplanetary Networking Preliminary Explorations and Empirical Results

Over the last decade interplanetary telecommunication capabilities have been significantly expanded--specifically in support of the Mars exploration rover and lander missions. NASA is continuing to drive advances in new, high payoff optical communications technologies to enhance the network to Gbps performance from Mars, and the transition from technology demonstration to operational system is examined through a hybrid RF/optical approach. Such a system combines the best features of RF and optical communications considering availability and performance to realize a dual band trunk line operating within characteristic constraints. Disconnection due to planetary obscuration and solar conjunction, link delays, timing, ground terminal mission congestion and scheduling policy along with space and atmospheric weather disruptions all imply the need for network protocol solutions to ultimately manage the physical layer in a transparent manner to the end user. Delay Tolerant Networking (DTN) is an approach under evaluation which addresses these challenges. A multi-hop multi-path hybrid RF and optical test bed has been constructed to emulate the integrated deep space network and to support protocol and hardware refinement. Initial experimental results characterize several of these challenges and evaluate the effectiveness of DTN as a solution to mitigate them.

Raible, Daniel E.↗

NASA NTRS2024

4K High Definition Video and Audio Streaming Across High-rate Delay Tolerant Space Networks

Audio and video streaming across delay tolerant networks are relatively new phenomena. During the Apollo 11 mission, video and audio were streamed directly back to Earth using fully analog radios. This streaming capability atrophied over time due to the gradual conversion to digital electronics accompanied with higher resolutions causing the required bit rates to outpace communication link performance. Additionally, 21st century space systems face the new requirement of interconnectedness. Delay Tolerant Networking (DTN) attempts to solve this requirement by uniting traditional point to point links into a robust and dynamic network. However, In order to avoid system bottlenecks, the High-Rate Delay Tolerant Networking (HDTN) implementation focuses on performance-optimization of the standards. This work extends the functionality of HDTN by implementing audio and video streaming, with the goal of demonstrating the practical application of high definition media streaming across space networks. A series of network topologies were created including simple point to point links and multi-node multi-hop networks. Test media in the form of prerecorded and live footage was streamed across the network. A set of objective quality metrics were established in order to measure the stream quality. A lunar network was emulated using a mixture of embedded ARM platforms.

Kyle J Vernyi↗

NASA NTRS2005

Airborne Subscale Transport Aircraft Research Testbed: Aircraft Model Development

The Airborne Subscale Transport Aircraft Research (AirSTAR) testbed being developed at NASA Langley Research Center is an experimental flight test capability for research experiments pertaining to dynamics modeling and control beyond the normal flight envelope. An integral part of that testbed is a 5.5% dynamically scaled, generic transport aircraft. This remotely piloted vehicle (RPV) is powered by twin turbine engines and includes a collection of sensors, actuators, navigation, and telemetry systems. The downlink for the plane includes over 70 data channels, plus video, at rates up to 250 Hz. Uplink commands for aircraft control include over 30 data channels. The dynamic scaling requirement, which includes dimensional, weight, inertial, actuator, and data rate scaling, presents distinctive challenges in both the mechanical and electrical design of the aircraft. Discussion of these requirements and their implications on the development of the aircraft along with risk mitigation strategies and training exercises are included here. Also described are the first training (non-research) flights of the airframe. Additional papers address the development of a mobile operations station and an emulation and integration laboratory.

Jordan, Thomas L.↗

NASA NTRS2019

Hazardous Product Detection and Environmental Clean-up Removal in Spacecraft Vehicles from Fire Induced Li-Ion Fires

A fire on-board the International Space Station (ISS) resulting from a commercial Surface Pro tablet Lithium-Ion (Li-ion) battery can be detrimental to the spacercraft and the astronauts. The Spacecraft Fire Safety Demonstration (Saffire) program is focused on identifying and quantifying the risks that potentially arise inside a space vehicle by conducting large fire experiments inside the Cygnus cargo vehicle upon re-entry to Earth. The potential candidate that will be flown on the ISS is a 4- cell Li-ion battery pack Surface Pro tablet (42 Wh). The tablets were tested at the White Sands Test Facility (WSTF) using a localized heating method to emulate the failing mechanism that leads the unit into thermal runaway. Measurements inside the test chamber were performed on aerosol mass concentrations and for specific toxic products (i.e. from CO, HCN, HCl and CO2). These toxic products depend on the size of the fire and energy content of the tablet. Comparisons will be made with the Dell XPS (97 Wh). The concentrations are then used to extrapolate to laptop fires on a vehicle at the approximate volumetric size of the Orion spacecraft. The presentation provides an analysis on the detection capability and the response time to trigger fire alarms aboard a vehicle by using the mass concentration levels. These results will consider the rate at which the life support system is able to filter the atmosphere in order to provide a hazardous free environment.

Padilla, Rosa E.↗

NASA NTRS2003

Physiological and Molecular Genetic Effects of Time-Varying Electromagnetic Fields on Human Neuronal Cells

The present investigation details the development of model systems for growing two- and three-dimensional human neural progenitor cells within a culture medium facilitated by a time-varying electromagnetic field (TVEMF). The cells and culture medium are contained within a two- or three-dimensional culture vessel, and the electromagnetic field is emitted from an electrode or coil. These studies further provide methods to promote neural tissue regeneration by means of culturing the neural cells in either configuration. Grown in two dimensions, neuronal cells extended longitudinally, forming tissue strands extending axially along and within electrodes comprising electrically conductive channels or guides through which a time-varying electrical current was conducted. In the three-dimensional aspect, exposure to TVEMF resulted in the development of three-dimensional aggregates, which emulated organized neural tissues. In both experimental configurations, the proliferation rate of the TVEMF cells was 2.5 to 4.0 times the rate of the non-waveform cells. Each of the experimental embodiments resulted in similar molecular genetic changes regarding the growth potential of the tissues as measured by gene chip analyses, which measured more than 10,000 human genes simultaneously.

Goodwin, Thomas J.↗

NASA NTRS2017

Model Validation for Propulsion - On the TFNS and LES Subgrid Models for a Bluff Body Stabilized Flame

With advances in computational power and availability of distributed computers, the use of even the most complex of turbulent chemical interaction models in combustors and coupled analysis of combustors and turbines is now possible and more and more affordable for realistic geometries. Recent more stringent emission standards have enticed the development of more fuel-efficient and low-emission combustion system for aircraft gas turbine applications. It is known that the NOx emissions tend to increase dramatically with increasing flame temperature. It is well known that the major difficulty, when modeling the turbulence-chemistry interaction, lies in the high non-linearity of the reaction rate expressed in terms of the temperature and species mass fractions. The transport filtered density function (FDF) model and the linear eddy model (LEM), which both use local instantaneous values of the temperature and mass fractions, have been shown to often provide more accurate results of turbulent combustion. In the present, the time-filtered Navier-Stokes (TFNS) approach capable of capturing unsteady flow structures important for turbulent mixing in the combustion chamber and two different subgrid models, LEM-like and EUPDF-like, capable of emulating the major processes occurring in the turbulence-chemistry interaction will be used to perform reacting flow simulations of a selected test case. The selected test case from the Volvo Validation Rig was documented by Sjunnesson.

bluff body↗

NASA NTRS1997

Test Report: Low-Cost Access to TDRS Using TOPEX to Emulate Small Satellite Performance

This report lists the objectives and conclusions of a series of experimental contacts between the TOPEX and the TDRS satellites. These experiments are designed to verify the theoretical prediction that a spin-stabilized satellite with a broad-beam, zenith-pointing antenna can have regular, significant contacts with the TDRS and use those contacts for data services. This series of experiments is a joint project between the experimenters at New Mexico State University (NMSU), the National Aeronautics and Space Administration (NASA) Goddard Space Flight Center (GSFC), and the Jet Propulsion Laboratory (JPL). In these experiments, we show that: (1) The satellite contacts during the experiment begin and end as predicted prior to the experiment; (2) The data contact is held for the desired contact duration; (3) The data quality through the contact is high and similar to that required by actual project needs; and (4) The receiving hardware at the White Sands Complex (WSC) is able to track the signals better than expected by analysis of the antenna pattern effects alone predict. We believe that these experiments successfully demonstrate the basic concept and its validity with actual spacecraft systems.

Horan, Stephen↗

NASA NTRS2012

Microhard MHX 2420 Orbital Performance Evaluation Using RT Logic T400CS

A major upfront cost of building low cost Nanosatellites is the communications sub-system. Most radios built for space missions cost over $4,000 per unit. This exceeds many budgets. One possible cost effective solution is the Microhard MHX2420, a commercial off-the-shelf transceiver with a unit cost under $1000. This paper aims to support the Nanosatellite community seeking an inexpensive radio by characterizing Microhard's performance envelope. Though not intended for space operations, the ability to test edge cases and increase average data transfer speeds through optimization positions this radio as a solution for Nanosatellite communications by expanding usage to include more missions. The second objective of this paper is to test and verify the optimal radio settings for the most common cases to improve downlinking. All tests were conducted with the aid of the RT Logic T400CS, a hardware-in-the-loop channel simulator designed to emulate real-world radio frequency (RF) link effects. This study provides recommended settings to optimize the downlink speed as well as the environmental parameters that cause the link to fail.

Microhard MHX 2420↗

NASA NTRS1987

Electrodeposition of metals and metal/cermet composites in low gravity

Electrodeposition experiments were carried out on the bench and a KC-135 aircraft at 0.01 g in anticipation of microgravity flights on NASA's Space Transportation System Shuttle. Experimental results obtained by interferometry compare concentration gradients as a function of time in the vicinity of a reducing electrode (cathode) for Cu(+2) and Co(+2) electrodeposition cells. No difference was found between bench and 0.01 g produced gradients for a .1M CuSO4 cell, but a significant difference was noted between the gradients in a 1M CoSO4 cell even though the bench cells were operated in a nonconvecting shielded (cathode over anode) mode. The gradient for Co(+2) depletion produced at 0.01 g was greater and the entire layer was thicker than found on the bench. Neutral buoyancy/matched density codeposition experiments were performed on the bench in an attempt to physically duplicate the results of metal/cermet codepositions in microgravity. Polystyrene spheres with average diameter 11.8 microns and density approximately matching that of 1M CoSO4 were utilized to emulate nonsedimenting cermets in microgravity. The cells were operated in a shielded convectionless mode. Comparison with literature data on codeposition with stirred cells indicate significant improvement in volume percent neutral occluded in the depositing metal matrix. A multicell electrodeposition flight apparatus that has been designed, constructed and is undergoing testing is discussed.

Riley, Clyde↗

NASA NTRS1993

JSC flight experiment recommendation in support of Space Station robotic operations

The man-tended configuration (MTC) of Space Station Freedom (SSF) provides a unique opportunity to move robotic systems from the laboratory into the mainstream space program. Restricted crew access due to the Shuttle's flight rate, as well as constrained on-orbit stay time, reduces the productivity of a facility dependent on astronauts to perform useful work. A natural tendency toward robotics to perform maintenance and routine tasks will be seen in efforts to increase SSF usefulness. This tendency will provide the foothold for deploying space robots. This paper outlines a flight experiment that will capitalize on the investment in robotic technology made by NASA over the past ten years. The flight experiment described herein provides the technology demonstration necessary for taking advantage of the expected opportunity at MTC. As a context to this flight experiment, a broader view of the strategy developed at the JSC is required. The JSC is building toward MTC by developing a ground-based SSF emulation funded jointly by internal funds, NASA/Code R, and NASA/Code M. The purpose of this ground-based Station is to provide a platform whereby technology originally developed at JPL, LaRC, and GSFC can be integrated into a near flight-like condition. For instance, the Automated Robotic Maintenance of Space Station (ARMSS) project integrates flat targets, surface inspection, and other JPL technologies into a Station analogy for evaluation. Also, ARMSS provides the experimental platform for the Capaciflector from GSPC to be evaluated for its usefulness in performing ORU change out or other tasks where proximity detection is required. The use and enhancement of these ground-based SSF models are planned for use through FY-93. The experimental data gathered from tests in these facilities will provide the basis for the technology content of the proposed flight experiment.

Berka, Reginald B.↗

NASA NTRS2014

NASA's Optical Measurement Program 2014

The Optical Measurements Group (OMG) within the NASA Orbital Debris Program Office (ODPO) addresses U.S. National Space Policy goals by monitoring and characterizing debris. Since 2001, the OMG has used the Michigan Orbital Debris Survey Telescope (MODEST) at Cerro Tololo Inter-American Observatory (CTIO) in Chile for general orbital debris surveys. The 0.6-m Schmidt MODEST provides calibrated astronomical data of GEO targets, both catalogued and uncatalogued debris, with excellent image quality. The data are utilized by the ODPO modeling group and are included in the Orbital Debris Engineering Model (ORDEM) v. 3.0. MODEST and the CTIO/SMARTS (Small and Moderate Aperture Research Telescope System) 0.9 m are both employed to acquire filter photometry data as well as synchronously observe targets in selected optical filters. Obtaining data synchronously yields data for material composition studies as well as longer orbital arc data on the same target without time delay or bias from a rotating, tumbling, or spinning target. Observations of GEO orbital debris using the twin 6.5-m Magellan telescopes at Las Campanas Observatory in Chile for deep imaging (Baade) and spectroscopic data (Clay) began in 2011. Through the data acquired on Baade, debris has been detected that reaches approx. 3 magnitudes fainter than detections with MODEST, while the spectral data from Clay provide better resolved information used in material characterization analyses. To better characterize and model optical data, the Optical Measurements Center (OMC) at NASA/JSC has been in operation since 2005, resulting in a database of comparison laboratory data. The OMC is designed to emulate illumination conditions in space using equipment and techniques that parallel telescopic observations and sourcetarget- sensor orientations. Lastly, the OMG is building the Meter Class Autonomous Telescope (MCAT) at Ascension Island. The 1.3-m telescope is designed to observe GEO and LEO targets, using a modified Ritchey-Chrétien configuration on a double horseshoe equatorial mount to allow tracking objects at LEO rates through the dome's keyhole at zenith. Through the data collection techniques employed at these unique facilities, NASA's ODPO has developed a multifaceted approach to characterize the orbital debris risk to satellites in various altitudes and provide insight leading toward material characterization of debris via photometric and spectroscopic measurements. Ultimately, the data are used in conjunction with in-situ and radar measurements to provide accurate data for models of our space environment and for facilitating spacecraft risk assessment.

Cowardin, H.↗

NASA NTRS2014

NASA's Optical Measurement Program 2014

The Optical Measurements Group (OMG) within the NASA Orbital Debris Program Office (ODPO) addresses U.S. National Space Policy goals by monitoring and characterizing debris. Since 2001, the OMG has used the Michigan Orbital Debris Survey Telescope (MODEST) at Cerro Tololo Inter-American Observatory (CTIO) in Chile for general orbital debris survey. The 0.6-m Schmidt MODEST provides calibrated astronomical data of GEO targets, both catalogued and uncatalogued debris, with excellent image quality. The data are utilized by the ODPO modeling group and are included in the Orbital Debris Engineering Model (ORDEM) v. 3.0. MODEST and the CTIO/SMARTS (Small and Moderate Aperture Research Telescope System) 0.9 m both acquire filter photometric data, as well as synchronously observing targets in selected optical filters. This information provides data used in material composition studies as well as longer orbital arc data on the same target, without time delay or bias from a rotating, tumbling, or spinning target. NASA, in collaboration with the University of Michigan, began using the twin 6.5-m Magellan telescopes at Las Campanas Observatory in Chile for deep imaging (Baade) and spectroscopic data (Clay) in 2011. Through the data acquired on Baade, debris have been detected that are ~3 magnitudes fainter than detections with MODEST, while the data from Clay provide better resolved information used in material characterization analyses via selected bandpasses. To better characterize and model optical data, the Optical Measurements Center (OMC) at NASA/JSC has been in operation since 2005, resulting in a database of comparison laboratory data. The OMC is designed to emulate illumination conditions in space using equipment and techniques that parallel telescopic observations and source-target-sensor orientations. Lastly, the OMG is building the Meter Class Autonomous Telescope (MCAT) at Ascension Island. The 1.3-m telescope is designed to observe GEO and LEO targets, using a modified Ritchey-Chrétien configuration on a double horseshoe equatorial mount to allow tracking objects at LEO rates through the dome's keyhole at zenith. Through the data collection techniques employed at these unique facilities, NASA's ODPO has developed a multi-faceted approach to characterize the orbital debris risk to satellites in various altitudes and provide material characterization of debris via photometric and spectroscopic measurements. Ultimately, the data are used in conjunction with in-situ and radar measurements to provide accurate data for models of our space environment and service spacecraft risk assessment.

Cowardin, H.↗

NASA NTRSDate not supplied

Hierarchical Control of Distributed Battery Energy Storage System in a DC Microgrid

This paper presents a novel hierarchical control approach of a DC microgrid (DCMG) which is supplied by a distributed battery energy storage system (BESS). With this approach, all battery units distributed in the BESS can be controlled to discharge with accurate current sharing and state-of-charge (SoC) balancing. Similar to other hierarchical control approaches used in DCMGs, this approach consists of three levels: (1) primary control, (2) secondary control and (3) tertiary control. This work includes defining a unit control error (UCE) at the secondary control level and evaluating current sharing weights at tertiary control level. A centralized controller at secondary control level is designed to detect the UCEs of each battery unit, and to restore the average voltage of a DCMG and control battery current sharing simultaneously. The distributed battery units share the load current in a DCMG based on weights. These weights are evaluated at the tertiary control level based on battery SoCs. The approach’s effectiveness was confirmed in digital simulation tests with the same simulation model as used in the NASA AMPS Modular Hardware Emulator.

Jing Zhang↗

NASA NTRS2019

Atmospheric Chemistry Modeling and Air Quality Forecasting Using Machine Learning

Atmospheric chemistry models are a central tool to study the impact of chemical constituents on the environment, vegetation and human health. These models split the atmosphere in a large number of grid-boxes and consider the emission of compounds into these boxes and their subsequent transport, deposition, and chemical processing. The chemistry is represented through a series of simultaneous ordinary differential equations, one for each compound. Given the difference in life-times between the chemical compounds (milli-seconds for O1D to years for CH4) these equations are numerically stiff and solving them consists of a significant fraction of the computational burden of a chemistry model.We have investigated a machine learning approach to emulate the chemistry instead of solving the differential equations numerically. From a one-month simulation of the GEOS-Chem model we have produced a training dataset consisting of the concentration of compounds before and after the differential equations are solved, together with some key physical parameters for every grid-box and time-step. From this dataset we have trained a machine learning algorithm (regression forest) to be able to predict the concentration of the compounds after the integration step based on the concentrations and physical state at the beginning of the time step. We have then included this algorithm back into the GEOS-Chem model, bypassing the need to integrate the chemistry.This machine learning approach shows many of the characteristics of the full simulation and has the potential to be substantially faster. There are a wide range of application for such an approach - generating boundary conditions, for use in air quality forecasts, chemical data assimilation systems, etc. We discuss speed and accuracy of our approach, and highlight some potential future directions for improving it.

Keller, Christoph A.↗

NASA NTRS2003

The development of inflatable array antennas

The deployable array antenna using an inflatable or thin-membrane structure has been identified as one of the enabling technologies to achieve low-mass, high packaging efficiency, low cost, and reliable deployment for future NASNJPL spacebome high-gain and large aperture antennas. Array antennas, when compared to parabolic reflectors, although suffering from limited bandwidth performance, offer wide-angle beam scanning capability and a more reliable flat “natural” aperture. To demonstrate the feasibility and capability of this low-mass array technology, three antenna concepts using inflatable and thin-membrane structures were initiated in 1997 at JPL and several breadboard units have been successfully developed. These three concepts are (1) the inflatable phased array, (2) the inflatable reflectarray, and (3) frame-supported thin membrane array antenna. All three concepts utilize the printed microstrip antenna technology. Although all three concepts will be briefly presented, this paper will discuss in more detail of the recent development of the second concept. The first concept of inflatable phased arrays that have been constructed are 3 to 5-meter size L-band dual-polarized synthetic aperture radar (SAR) arrays for Earth remote sensing application. They all consist of a rectangular configuration with inflatable cylindrical tubes that support and tension a multi-layer thin-membrane radiating aperture with microstrip patches and microstrip power divider lines. For the second concept, an inflatable reflectarray was developed for future deep-space telecom applications. It is a 3m Ka-band inflatable reflectarray (1.8 kg/m2). This reflectarray uses inflated torus tubes to support and tension a flat-membrane reflectarray surface. The reflectarray surface emulates a curved parabolic reflecting surface. However, because of its flat surface being a “natural surface”, it is much easier to achieve and more reliable to maintain the required surface tolerance than that of a curved parabola during long space flight. For the third concept, an L-band dual polarized SAR array with a 5m x 3m aperture and 2.5 kg/m2 of mass has recently been demonstrated. It consists of seven foldable panels each having a rectangular frame that supports a two-layer thinmembrane microstrip subarray aperture. Each frame is made of light-weight graphite composite material. The chief advantage of this deployable “frame” concept is that each frame is able to rigidly support an appropriate number of T/R modules and phase shifters to achieve the desired power distribution and beam scanning. Several technology challenges, such as the development of rigidizable inflatable tubes, a controlled deployment mechanism, thin-membrane thermal effects, a low-mass inflation system, membrane mounted T/R modules, means to counter surface tolerance issues, etc. are being investigated and will be discussed in the presentation. With foreseeable success in the development of these challenging areas, the inflatable/thin-membrane array antennas could be mature enough in a few years for actual space flight.

Lou, Michael↗

NASA NTRS2019

Atmospheric Chemistry Modeling Using Machine Learning

Atmospheric chemistry models are a central tool to study the impact of chemical constituents on the environment, vegetation and human health. These models split the atmosphere in a large number of grid-boxes and consider the emission of compounds into these boxes and their subsequent transport, deposition, and chemical processing. The chemistry is represented through a series of simultaneous ordinary differential equations, one for each compound. Given the difference in life-times between the chemical compounds (milli-seconds for O1D to years for CH4) these equations are numerically stiff and solving them consists of a significant fraction of the computational burden of a chemistry model. We have investigated a machine learning approach to emulate the chemistry instead of solving the differential equations numerically. From a one-month simulation of the GEOS-Chem model we have produced a training dataset consisting of the concentration of compounds before and after the differential equations are solved, together with some key physical parameters for every grid-box and time-step. From this dataset we have trained a machine learning algorithm (regression forest) to be able to predict the concentration of the compounds after the integration step based on the concentrations and physical state at the beginning of the time step. We have then included this algorithm back into the GEOS-Chem model, bypassing the need to integrate the chemistry. This machine learning approach shows many of the characteristics of the full simulation and has the potential to be substantially faster. There are a wide range of application for such an approach - generating boundary conditions, for use in air quality forecasts, chemical data assimilation systems, etc. We discuss speed and accuracy of our approach, and highlight some potential future directions for improving it.

Keller, Christoph A.↗