Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Predictive 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.

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

At least 1,171 records · Page 65
All sourcesNASA NTRSDOE OSTI
NASA NTRS2012

An Efficient Deterministic Approach to Model-based Prediction Uncertainty Estimation

Prognostics deals with the prediction of the end of life (EOL) of a system. EOL is a random variable, due to the presence of process noise and uncertainty in the future inputs to the system. Prognostics algorithm must account for this inherent uncertainty. In addition, these algorithms never know exactly the state of the system at the desired time of prediction, or the exact model describing the future evolution of the system, accumulating additional uncertainty into the predicted EOL. Prediction algorithms that do not account for these sources of uncertainty are misrepresenting the EOL and can lead to poor decisions based on their results. In this paper, we explore the impact of uncertainty in the prediction problem. We develop a general model-based prediction algorithm that incorporates these sources of uncertainty, and propose a novel approach to efficiently handle uncertainty in the future input trajectories of a system by using the unscented transformation. Using this approach, we are not only able to reduce the computational load but also estimate the bounds of uncertainty in a deterministic manner, which can be useful to consider during decision-making. Using a lithium-ion battery as a case study, we perform several simulation-based experiments to explore these issues, and validate the overall approach using experimental data from a battery testbed.

Daigle, Matthew J.↗
DOE OSTI2018

Combination of searches for heavy resonances decaying into bosonic and leptonic final states using 36 fb - 1 of proton-proton collision data at s = 13 TeV with the ATLAS detector

Searches for new heavy resonances decaying into different pairings of W, Z, or Higgs bosons, as well as directly into leptons, are presented using a data sample corresponding to 36.1 fb -1 of pp collisions at $ \sqrt{s}=13 $ TeV collected during 2015 and 2016 with the ATLAS detector at the CERN Large Hadron Collider. Analyses selecting bosonic decay modes in the qqqq, ννqq, ℓνqq, ℓℓqq, ℓνℓν, ℓℓνν, ℓνℓℓ, ℓℓℓℓ, qqbb, ννbb, ℓνbb, and ℓℓbb final states are combined, searching for a narrow-width resonance. Likewise, analyses selecting the leptonic ℓν and ℓℓ final states are also combined. These two sets of analyses are then further combined. No significant deviation from the Standard Model predictions is observed. Three benchmark models are tested: a model predicting the existence of a new heavy scalar singlet, a simplified model predicting a heavy vector-boson triplet, and a bulk Randall-Sundrum model with a heavy spin-2 Kaluza-Klein excitation of the graviton. Cross section limits are set at the 95% confidence level using an asymptotic approximation and are compared with predictions for the benchmark models. These limits are also expressed in terms of constraints on couplings of the heavy vector-boson triplet to quarks, leptons, and the Higgs boson. The data exclude a heavy vector-boson triplet with mass below 5.5 TeV in a weakly coupled scenario and 4.5 TeV in a strongly coupled scenario, as well as a Kaluza-Klein graviton with mass below 2.3 TeV.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2016

Searches for heavy diboson resonances in pp collisions at $ \sqrt{s}=13 $ TeV with the ATLAS detector

Searches for new heavy resonances decaying to WW, WZ, and ZZ bosons are presented, using a data sample corresponding to 3.2 fb -1 of pp collisions at s=13 TeV colle cted with the ATLAS detector at the CERN Large Hadron Collider. Analyses selecting ννqq, ℓνqq, ℓℓqq and qqqq final states are combined, searching for an arrow-width resonance with mass between 500 and 3000 GeV. The discriminating variable is either an invariant mass or a transverse mass. No significant deviations from the Standard Model predictions are observed. Three benchmark models are tested: a model predicting the existence of a new heavy scalar singlet, a simplified model predicting a heavy vector-boson triplet, and a bulk Randall-Sundrum model with a heavy spin-2 graviton. Cross-section limits are set at the 95% confidence level and are compared to theoretical cross-section predictions for a variety of models. The data exclude a scalar singlet with mass below 2650 GeV, a heavy vector-boson triplet with mass below 2600 GeV, and a graviton with mass below 1100 GeV. These results significantly extend the previous limits set using pp collisions at √s=8 TeV.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS2012

Impact of Land Model Calibration on Coupled Land-Atmosphere Prediction

Land-atmosphere (L-A) interactions play a critical role in determining the diurnal evolution of both planetary boundary layer (PBL) and land surface heat and moisture budgets, as well as controlling feedbacks with clouds and precipitation that lead to the persistence of dry and wet regimes. Recent efforts to quantify the strength of L-A coupling in prediction models have produced diagnostics that integrate across both the land and PBL components of the system. In this study, we examine the impact of improved specification of land surface states, anomalies, and fluxes on coupled WRF forecasts during the summers of extreme dry and wet land surface conditions in the U.S. Southern Great Plains. The improved land initialization and surface flux parameterizations are obtained through calibration of the Noah land surface model using the new optimization and uncertainty estimation subsystem in NASA's Land Information System (LIS-OPT/UE). The impact of the calibration on the a) spinup of the land surface used as initial conditions, and b) the simulated heat and moisture states and fluxes of the coupled WRF simulations is then assessed. Changes in ambient weather and land-atmosphere coupling are evaluated along with measures of uncertainty propagation into the forecasts. In addition, the sensitivity of this approach to the period of calibration (dry, wet, average) is investigated. Results indicate that the offline calibration leads to systematic improvements in land-PBL fluxes and near-surface temperature and humidity, and in the process provide guidance on the questions of what, how, and when to calibrate land surface models for coupled model prediction.

Santanello, Joseph A., Jr.↗
DOE OSTI2019

Aluminum cladding oxide growth prediction for high flux research reactors

Aluminum cladding oxidation of research-reactor fuel elements at high power conditions has a disadvantageous effect on fuel performance due to the lower thermal conductivity of the oxide. The oxide growth prediction models available in the literature were mostly developed for low power conditions. To examine the applicability of the models to high power and high temperature test conditions, the models were studied by coupling with the most frequently employed heat transfer coefficient (HTC) correlations including the Dittus-Boelter correlation, the Colburn correlation, the Sieder-Tate correlation, and KAERI-developed correlation. The Griess model over-predicted the oxide growth while the KAERI-Griess model under-predicted the oxide growth for high power tests. The Kim model, coupled with the Colburn correlation, gave most consistent results with the measured data from two BR2 experiments. However, the Kim model was found to be inapplicable to the EUHFRR conditions at the peak power locations if it was coupled with the Dittus-Boelter correlation. A revision of the prediction models to more closely agree with the measured data was recommended. Furthermore, AG3NE and AlFeNi cladding types were tested in the E-FUTURE experiment, and a noticeable (although small) reduction in oxide thickness on the AlFeNi cladding was observed. However, this difference was believed to be only a secondary effect considering other uncertainties in model predictions, so no attempt was made to model the alloying effect.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2020

Development and Evaluation of Occupancy-Aware HVAC Control for Residential Building Energy Efficiency and Occupant Comfort

Occupancy-aware heating, ventilation, and air conditioning (HVAC) control offers the opportunity to reduce energy use without sacrificing thermal comfort. Residential HVAC systems often use manually-adjusted or constant setpoint temperatures, which heat and cool the house regardless of whether it is needed. By incorporating occupancy-awareness into HVAC control, heating and cooling can be used for only those time periods it is needed. Yet, bringing this technology to fruition is dependent on accurately predicting occupancy. Non-probabilistic prediction models offer an opportunity to use collected occupancy data to predict future occupancy profiles. Smart devices, such as a connected thermostat, which already include occupancy sensors, can be used to provide a continually growing collection of data that can then be harnessed for short-term occupancy prediction by compiling and creating a binary occupancy prediction. Real occupancy data from six homes located in Colorado is analyzed and investigated using this occupancy prediction model. Results show that non-probabilistic occupancy models in combination with occupancy sensors can be combined to provide a hybrid HVAC control with savings on average of 5.0% and without degradation of thermal comfort. Model predictive control provides further opportunities, with the ability to adjust the relative importance between thermal comfort and energy savings to achieve savings between 1% and 13.3% depending on the relative weighting between thermal comfort and energy savings. In all cases, occupancy prediction allows the opportunity for a more intelligent and optimized strategy to residential HVAC control.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS1973

Mathematical model for predicting human vertebral fracture

Mathematical model has been constructed to predict dynamic response of tapered, curved beam columns in as much as human spine closely resembles this form. Model takes into consideration effects of impact force, mass distribution, and material properties. Solutions were verified by dynamic tests on curved, tapered, elastic polyethylene beam.

Benedict, J. V.↗
DOE OSTI2022

Liquid Core Detection and Strand Condition Monitoring in a Continuous Caster Using Optical Fiber

Real-time monitoring of the liquid core position during the continuous casting of steel has been demonstrated using low-cost distributed optical-fiber-based strain sensors. These sensors were installed on the containment roll support structures in the segments of a production continuous caster to detect the position of the solid–liquid interface and monitor the strand condition during the continuous casting. Distributed Fiber Bragg Grating sensors (FBGs) were used in this work to monitor strain at six roll positions in the caster. The sensor performance was first validated by comparing optical strain measurements with conventional strain gauge measurements in the lab. Next, optical strain measurements were performed on an isolated caster segment in a segment maintenance facility using hydraulic jacks to simulate the presence of a liquid core under the roll. Finally, the sensors were evaluated during caster operation. The sensors successfully detected the load increase associated with the presence of a liquid core under each instrumented roll location. Incidents of bulging and roll eccentricity were also detected using frequency analysis of the optical strain signal. The liquid core position measurements were compared using predictions from computer models (digital twins) in use at the production site. The measurements were in good agreement with the model predictions, with a few exceptions. Under certain transient caster operating conditions, such as spraying practice changes and SEN exchanges, the model predictions deviated slightly from the liquid core position determined from strain measurements.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Mean Climate and Tropical Rainfall Variability in Aquaplanet Simulations Using the Model for Prediction Across Scales‐Atmosphere

Abstract Aquaplanet experiments are important tools for understanding and improving physical processes simulated by global models; yet, previous aquaplanet experiments largely differ in their representation of subseasonal tropical rainfall variability. This study presents results from aquaplanet experiments produced with the Model for Prediction Across Scales‐Atmosphere (MPAS‐A)—a community model specifically designed to study weather and climate in a common framework. The mean climate and tropical rainfall variability simulated by MPAS‐A with varying horizontal resolution were compared against results from a recent suite of aquaplanet experiments. This comparison shows that, regardless of horizontal resolution, MPAS‐A produces the expected mean climate of an aquaplanet framework with zonally symmetric but meridionally varying sea‐surface temperature. MPAS‐A, however, has a stronger signal of tropical rainfall variability driven by convectively coupled equatorial waves. Sensitivity experiments with different cumulus parameterizations, physics packages, and vertical grids consistently show the presence of those waves, especially equatorial Kelvin waves, in phase with lower‐tropospheric convergence. Other models do not capture such rainfall‐kinematics phasing. These results suggest that simulated tropical rainfall variability depends not only on the cumulus parameterization (as suggested by previous studies) but also on the coupling between physics and dynamics of climate and weather prediction models.

Rios‐Berrios, R.↗
DOE OSTI2015

Modeling the effectiveness of remediation efforts in contaminated urban areas: An EMRAS II Urban Areas Working Group exercise - 15631

The Urban Areas Working Group was part of the IAEA's EMRAS II (Environmental Modelling for Radiation Safety) Programme during 2009-2011. The Working Group's goal was to test and improve the capabilities of models used in assessment of radioactive contamination in urban settings. The 'contaminant transport and countermeasures' exercise was focused on modeling the effectiveness of potential countermeasures or remediation efforts with respect to reduction of human exposures and doses. The exercise was based on a hypothetical situation involving an initial concentration of Co-60 or Pu-239 in air in a specified urban area. Participants were given detailed input information, including descriptions of test locations in two types of urban areas, (1) a business area with buildings and asphalt, and (2) a park area near an apartment town. Several kinds of initial weather conditions (dry, light rain, and heavy rain) were explored. Modeling endpoints included contamination densities, dose rates (from each contributing surface and total), annual and cumulative doses (internal and external) for specified reference individuals, and effectiveness of countermeasures and remediation efforts in terms of dose rate reduction and dose reduction. External doses were emphasized for Co-60 and internal doses for Pu-239. Approaches and model predictions from five models are compared, and explanations for similarities and differences among model predictions are offered. Understanding the effects of different approaches, assumptions, and parameter values is essential to a comparison of results from different models and to understanding both the usefulness and the limitations of any given model. (authors)

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2008

On Convergence of Development Costs and Cost Models for Complex Spaceflight Instrument Electronics

Development costs of a few recent spaceflight instrument electrical and electronics subsystems have diverged from respective heritage cost model predictions. The cost models used are Grass Roots, Price-H and Parametric Model. These cost models originated in the military and industry around 1970 and were successfully adopted and patched by NASA on a mission-by-mission basis for years. However, the complexity of new instruments recently changed rapidly by orders of magnitude. This is most obvious in the complexity of representative spaceflight instrument electronics' data system. It is now required to perform intermediate processing of digitized data apart from conventional processing of science phenomenon signals from multiple detectors. This involves on-board instrument formatting of computational operands from row data for example, images), multi-million operations per second on large volumes of data in reconfigurable hardware (in addition to processing on a general purpose imbedded or standalone instrument flight computer), as well as making decisions for on-board system adaptation and resource reconfiguration. The instrument data system is now tasked to perform more functions, such as forming packets and instrument-level data compression of more than one data stream, which are traditionally performed by the spacecraft command and data handling system. It is furthermore required that the electronics box for new complex instruments is developed for one-digit watt power consumption, small size and that it is light-weight, and delivers super-computing capabilities. The conflict between the actual development cost of newer complex instruments and its electronics components' heritage cost model predictions seems to be irreconcilable. This conflict and an approach to its resolution are addressed in this paper by determining the complexity parameters, complexity index, and their use in enhanced cost model.

Kizhner, Semion↗
DOE OSTI2016

Status of Chronic Oxidation Studies of Graphite

Graphite will undergo extremely slow, but continuous oxidation by traces of moisture that will be present, albeit at very low levels, in the helium coolant of HTGR. This chronic oxidation may cause degradation of mechanical strength and thermal properties of graphite components if a porous oxidation layer penetrates deep enough in the bulk of graphite components during the lifetime of the reactor. The current research on graphite chronic oxidation is motivated by the acute need to understand the behavior of each graphite grade during prolonged exposure to high temperature chemical attack by moisture. The goal is to provide the elements needed to develop predictive models for long-time oxidation behavior of graphite components in the cooling helium of HTGR. The tasks derived from this goal are: (1) Oxidation rate measurements in order to determine and validate a comprehensive kinetic model suitable for prediction of intrinsic oxidation rates as a function of temperature and oxidant gas composition; (2) Characterization of effective diffusivity of water vapor in the graphite pore system in order to account for the in-pore transport of moisture; and (3) Development and validation of a predictive model for the penetration depth of the oxidized layer, in order to assess the risk of oxidation caused damage of particular graphite grades after prolonged exposure to the environment of helium coolant in HTGR. The most important and most time consuming of these tasks is the measurement of oxidation rates in accelerated oxidation tests (but still under kinetic control) and the development of a reliable kinetic model. This report summarizes the status of chronic oxidation studies on graphite, and then focuses on model development activities, progress of kinetic measurements, validation of results, and improvement of the kinetic models. Analysis of current and past results obtained with three grades of showed that the classical Langmuir-Hinshelwood model cannot reproduce all data collected so far. Starting from here we propose a modification of the LH model to include temperature activation of graphite surface as a Boltzmann activation function. The enhanced Boltzmann-Langmuir-Hinshelwood model (BLH) was tested successfully on three grades of graphite. The model is a robust, comprehensive mathematical function that allows better fitting of experimental results spanning a wide range of temperature and partial pressures of water vapor and hydrogen. However, the model did not fit satisfactorily the data extracted from the old report on graphite H-451 oxidation by water.

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

Integrated Computational Materials and Mechanical Modeling for Additive Manufacturing of Alloys with Graded Structure Used in Fossil Fuel Power Plants

Wire-arc additive manufacturing (WAAM) has demonstrated its unique capability of producing large-size alloy components with a significantly reduced fabrication time and enhanced geometry design freedom. In this project, the team has developed an ICME (Integrated Computational Materials Engineering) modeling framework, which supports the WAAM of the AUSC (Advanced Ultra-Supercritical) power plant components. The manufacturing design has been applied to Inconel 740H, steel P91, as well as the dissimilar alloy components between steel P91 and Inconel 740H. The ICME model framework is developed by considering two types of modeling. First, mechanistic modeling has been applied to control the printing quality and understand the sequence of the dissimilar printing of the wall structure. The following models have been included in the developed ICME framework: finite element thermal model, grain structure model, residual stress simulation, crystal plasticity model, CALPHAD-based precipitation kinetic model, phase stability prediction, thermal expansion predictive model, and heuristic creep model. Secondary, a physics-based machine learning model has also been developed based on the ICME model structure. The machine learning model development is based on the ICME model prediction with calibration of the experiments. In addition, the WAAM has been utilized as a high-throughput experimental tool rapidly generating a gradient of alloy composition to facilitate experimental database generation for process-structure-property relationships. Such a database directly supported the ICME-enhanced machine learning, which further assisted in intermediate composition block design between P91 and 740H. A high-throughput screening study of the oxidation resistance has been performed based on such high-throughput experimentation. Based on the computational design, several dissimilar alloy manufacturing with post-heat treatment have been performed with a comprehensive evaluation of mechanical performance, including hardness mapping, yield strength, creep resistance. In this project, the single component of P91 and 740H processed by WAAM after heat treatment designed by ICME has demonstrated higher performance in yield strength and creep resistance than the wrought materials. The P91 sample prepared by WAAM with ICME-designed heat treatment performs better than P92 in creep resistance. The designed graded alloy printing with intermediate block shows a promising performance that exceeds the traditional welding. Moreover, the current research indicates the high need for location-specific design analysis with uncertainty quantification, an important topic that deserves more dedicated research. The achievement of this project demonstrated the promising future of WAAM in structural alloy manufacturing for energy power plant development. Successful printing requires synergetic efforts made by manufacturing, mechanical, and materials sciences.

20 FOSSIL-FUELED POWER PLANTS↗
NASA NTRS1978

Evaluation of ceramics for stator application: Gas turbine engine report

Current ceramic materials, component fabrication processes, and reliability prediction capability for ceramic stators in an automotive gas turbine engine environment are assessed. Simulated engine duty cycle testing of stators conducted at temperatures up to 1093 C is discussed. Materials evaluated are SiC and Si3N4 fabricated from two near-net-shape processes: slip casting and injection molding. Stators for durability cycle evaluation and test specimens for material property characterization, and reliability prediction model prepared to predict stator performance in the simulated engine environment are considered. The status and description of the work performed for the reliability prediction modeling, stator fabrication, material property characterization, and ceramic stator evaluation efforts are reported.

Trela, W.↗
NASA NTRS1980

Status of the DOE/NASA critical gas turbine research and technology project

Activities performed in order to provide an R&T data base for utility gas turbine systems burning coal-derived fuels are described. Experiments were run to determine the corrosivity effects of trace metal contaminants (and potential fuel additives) on gas turbine materials and these results were correlated in a corrosion-life prediction model. Actual fuels were burned in a burner rig hot corrosion test to verify the model. A deposition prediction model was assembled and compared with results of actual coal-derived fuel deposition tests. Thermal barrier coatings were tested to determine their potential for protecting gas turbine hardware from the corrosive contaminants. Several coatings were identified with significantly improved spallation-resistance (and, hence, corrosion resistance).

Clark, J. S.↗
DOE OSTI2020

LA17-V-InverseRadTransportProblem-PD2Kb. Final Project Report

In the inverse radiation transport problem, a radiation signal emanating from an unknown object is used to reconstruct geometric and material parameters of the object , e.g., the mass of special nuclear material (SNM), the isotopic composition of the SNM (e.g., the enrichment of 235U), the thickness of shielding, etc. The inverse radiation transport problem is one of the most fundamental and important in the emergency response mission, and has been addressed in previous NA-22 and NA-42/84 investments by using derivative-free optimization methods. Our project was aimed at (i) consolidating previous work by applying polynomial chaos expansion (PCE) for reducing significantly the number of forward model simulations required by derivative-free optimization methods; (ii) developing a general-purpose framework and stand-alone software for solving the inverse radiation transport problem by applying a new “predictive modeling” methodology (predictive modeling of coupled multi-physics systems, PM_CMPS) which enables the fusing of experimental and computational information; and (iii) developing several benchmarks (for unscattered gamma rays) for demonstrating the quantification of non-Gaussian features (e.g., skewness and kurtosis, which characterize asymmetries and long tails) of response distributions by deterministically and efficiently propagating parameter uncertainties (covariances and higher-order moments, where available) using high-order sensitivities computed by the recently developed comprehensive adjoint sensitivity analysis methodology (C-ASAM). With regard to (i), application of PCE, the most significant accomplishment of this project was an implementation of PCE into INVERSE that significantly reduced computation time for the DiffeRential Evolution Adaptive Metropolis (DREAM) and Differential Evolution (DE) methods. For inverse transport problems with measurements of neutron leakage, neutron multiplication, and neutron-induced gamma-rays, the solution times for the DREAM and DE methods were reduced by factors of 30-300, reducing run times from hours to minutes while maintaining the same accuracy as the standard (non-PCE) DREAM and DE. Speedups were further improved by exploring sparse grid quadrature sets using the Toolkit for Adaptive Stochastic Modeling and Non-Intrusive ApproximatioN (Tasmanian) toolset. An initial exploration of sparse grids indicated that further speedups factors of 5-10 are obtainable. The implementation of PCE in INVERSE is not in a form that users can easily access. With regard to (ii), solving the inverse problem using a predictive modeling paradigm, the most significant accomplishment of this project were the derivation and implementation of adjoint-based second derivatives in neutron and photon transport theory. There are two codes that compute these derivatives. SENSPG computes them for uncollided gamma rays and SENSMG computes them for neutron leakage. SENSMG is open-source code available on github. Other significant accomplishments included a stand-alone code, MULTI-PRED, that implements the predictive modeling paradigm, along with a detailed manual. The code was implemented in the existing INVERSE code for inverse problems, but a head-to-head comparison of inverse methods was not achieved. We also derived and implemented the capability to compute sensitivities for multiplicity measurements into the SENSMG code, and this capability was used in a data assimilation application, but it was not used for solving inverse problems. With regard to (iii), developing benchmarks to demonstrate the importance of high-order sensitivity and uncertainty analysis, the most significant accomplishment of this project were two test problems that showed how important second-order sensitivities are.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2023

Machine learning models inaccurately predict current and future high-latitude C balances

The high-latitude carbon (C) cycle is a key feedback to the global climate system, yet because of system complexity and data limitations, there is currently disagreement over whether the region is a source or sink of C. Recent advances in big data analytics and computing power have popularized the use of machine learning (ML) algorithms to upscale site measurements of ecosystem processes, and in some cases forecast the response of these processes to climate change. Due to data limitations, however, ML model predictions of these processes are almost never validated with independent datasets. To better understand and characterize the limitations of these methods, we develop an approach to independently evaluate ML upscaling and forecasting. We mimic data-driven upscaling and forecasting efforts by applying ML algorithms to different subsets of regional process-model simulation gridcells, and then test ML performance using the remaining gridcells. In this study, we simulate C fluxes and environmental data across Alaska using ecosys, a process-rich terrestrial ecosystem model, and then apply boosted regression tree ML algorithms to training data configurations that mirror and expand upon existing AmeriFLUX eddy-covariance data availability. We first show that a ML model trained using ecosys outputs from currently-available Alaska AmeriFLUX sites incorrectly predicts that Alaska is presently a modeled net C source. Increased spatial coverage of the training dataset improves ML predictions, halving the bias when 240 modeled sites are used instead of 15. However, even this more accurate ML model incorrectly predicts Alaska C fluxes under 21st century climate change because of changes in atmospheric CO 2 , litter inputs, and vegetation composition that have impacts on C fluxes which cannot be inferred from the training data. Our results provide key insights to future C flux upscaling efforts and expose the potential for inaccurate ML upscaling and forecasting of high-latitude C cycle dynamics.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2000

Predictive Multiple Model Switching Control with the Self-Organizing Map

A predictive, multiple model control strategy is developed by extension of self-organizing map (SOM) local dynamic modeling of nonlinear autonomous systems to a control framework. Multiple SOMs collectively model the global response of a nonautonomous system to a finite set of representative prototype controls. Each SOM provides a codebook representation of the dynamics corresponding to a prototype control. Different dynamic regimes are organized into topological neighborhoods where the adjacent entries in the codebook represent the global minimization of a similarity metric. The SOM is additionally employed to identify the local dynamical regime, and consequently implements a switching scheme that selects the best available model for the applied control. SOM based linear models are used to predict the response to a larger family of control sequences which are clustered on the representative prototypes. The control sequence which corresponds to the prediction that best satisfies the requirements on the system output is applied as the external driving signal.

Motter, Mark A.↗