Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “inverse model”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS2011

Hybrid Adaptive Flight Control with Model Inversion Adaptation

This study investigates a hybrid adaptive flight control method as a design possibility for a flight control system that can enable an effective adaptation strategy to deal with off-nominal flight conditions. The hybrid adaptive control blends both direct and indirect adaptive control in a model inversion flight control architecture. The blending of both direct and indirect adaptive control provides a much more flexible and effective adaptive flight control architecture than that with either direct or indirect adaptive control alone. The indirect adaptive control is used to update the model inversion controller by an on-line parameter estimation of uncertain plant dynamics based on two methods. The first parameter estimation method is an indirect adaptive law based on the Lyapunov theory, and the second method is a recursive least-squares indirect adaptive law. The model inversion controller is therefore made to adapt to changes in the plant dynamics due to uncertainty. As a result, the modeling error is reduced that directly leads to a decrease in the tracking error. In conjunction with the indirect adaptive control that updates the model inversion controller, a direct adaptive control is implemented as an augmented command to further reduce any residual tracking error that is not entirely eliminated by the indirect adaptive control.

Nguyen, Nhan↗
DOE OSTI2022

Inverse Modeling of Hydrologic Parameters in CLM4 via Generalized Polynomial Chaos in the Bayesian Framework

In this work, generalized polynomial chaos (gPC) expansion for land surface model parameter estimation is evaluated. We perform inverse modeling and compute the posterior distribution of the critical hydrological parameters that are subject to great uncertainty in the Community Land Model (CLM) for a given value of the output LH. The unknown parameters include those that have been identified as the most influential factors on the simulations of surface and subsurface runoff, latent and sensible heat fluxes, and soil moisture in CLM4.0. We set up the inversion problem in the Bayesian framework in two steps: (i) building a surrogate model expressing the input–output mapping, and (ii) performing inverse modeling and computing the posterior distributions of the input parameters using observation data for a given value of the output LH. The development of the surrogate model is carried out with a Bayesian procedure based on the variable selection methods that use gPC expansions. Our approach accounts for bases selection uncertainty and quantifies the importance of the gPC terms, and, hence, all of the input parameters, via the associated posterior probabilities.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2013

Inverse Modeling of Texas NOx Emissions Using Space-Based and Ground-Based NO2 Observations

Inverse modeling of nitrogen oxide (NOx) emissions using satellite-based NO2 observations has become more prevalent in recent years, but has rarely been applied to regulatory modeling at regional scales. In this study, OMI satellite observations of NO2 column densities are used to conduct inverse modeling of NOx emission inventories for two Texas State Implementation Plan (SIP) modeling episodes. Addition of lightning, aircraft, and soil NOx emissions to the regulatory inventory narrowed but did not close the gap between modeled and satellite observed NO2 over rural regions. Satellitebased top-down emission inventories are created with the regional Comprehensive Air Quality Model with extensions (CAMx) using two techniques: the direct scaling method and discrete Kalman filter (DKF) with Decoupled Direct Method (DDM) sensitivity analysis. The simulations with satellite-inverted inventories are compared to the modeling results using the a priori inventory as well as an inventory created by a ground-level NO2 based DKF inversion. The DKF inversions yield conflicting results: the satellite based inversion scales up the a priori NOx emissions in most regions by factors of 1.02 to 1.84, leading to 3-55% increase in modeled NO2 column densities and 1-7 ppb increase in ground 8 h ozone concentrations, while the ground-based inversion indicates the a priori NOx emissions should be scaled by factors of 0.34 to 0.57 in each region. However, none of the inversions improve the model performance in simulating aircraft-observed NO2 or ground-level ozone (O3) concentrations.

Nitrogen dioxide↗
DOE OSTI2024

Uncertainty Quantification in GADRAS Inverse Modeling

The Gamma Detector Response and Analysis Software (GADRAS) package includes an inverse modeling tool that is helpful in identifying characteristics of unknown radioactive materials. Traditionally, uncertainties in this analysis were derived solely from measurement data quality and the fit of synthetic spectra. This paper aims to rigorously quantify additional sources of uncertainty, focusing on uncertainties arising from measurements being analyzed, Detector Response Function (DRF) characterization, and DRF extrapolation. Applying these findings to the BeRPBall benchmark data set, we demonstrated the impact of these uncertainties on plutonium and polyethylene estimates. The results underscore the importance of incorporating diverse uncertainty sources to enhance the accuracy and reliability of GADRAS’s inverse modeling capabilities.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS2012

Sensitivity of Ocean Reflectance Inversion Models for Identifying and Discriminating Between Phytoplankton Functional Groups

The daily, synoptic images provided by satellite ocean color instruments provide viable data streams for observing changes in the biogeochemistrY of marine ecosystems. Ocean reflectance inversion models (ORMs) provide a common mechanism for inverting the "color" of the water observed a satellite into marine inherent optical properties (lOPs) through a combination of empiricism and radiative transfer theory. lOPs, namely the spectral absorption and scattering characteristics of ocean water and its dissolved and particulate constituents, describe the contents of the upper ocean, information critical for furthering scientific understanding of biogeochemical oceanic processes. Many recent studies inferred marine particle sizes and discriminated between phytoplankton functional groups using remotely-sensed lOPs. While all demonstrated the viability of their approaches, few described the vertical distributions of the water column constituents under consideration and, thus, failed to report the biophysical conditions under which their model performed (e.g., the depth and thickness of the phytoplankton bloom(s)). We developed an ORM to remotely identifY Noctiluca miliaris and other phytoplankton functional types using satellite ocean color data records collected in the northern Arabian Sea. Here, we present results from analyses designed to evaluate the applicability and sensitivity of the ORM to varied biophysical conditions. Specifically, we: (1) synthesized a series of vertical profiles of spectral inherent optical properties that represent a wide variety of bio-optical conditions for the northern Arabian Sea under aN Miliaris bloom; (2) generated spectral remote-sensing reflectances from these profiles using Hydrolight; and, (3) applied the ORM to the synthesized reflectances to estimate the relative concentrations of diatoms and N Miliaris for each example. By comparing the estimates from the inversion model to those from synthesized vertical profiles, we were able to identifY those bio-optical conditions under which the inversion model performs both well and poorly.

Werdell, P. Jeremy↗
DOE OSTI2013

Impact of transport model errors on the global and regional methane emissions estimated by inverse modelling

A modelling experiment has been conceived to assess the impact of transport model errors on methane emissions estimated in an atmospheric inversion system. Synthetic methane observations, obtained from 10 different model outputs from the international TransCom-CH4 model inter-comparison exercise, are combined with a prior scenario of methane emissions and sinks, and integrated into the three-component PYVAR-LMDZ-SACS (PYthon VARiational-Laboratoire de Météorologie Dynamique model with Zooming capability-Simplified Atmospheric Chemistry System) inversion system to produce 10 different methane emission estimates at the global scale for the year 2005. The same methane sinks, emissions and initial conditions have been applied to produce the 10 synthetic observation datasets. The same inversion set-up (statistical errors, prior emissions, inverse procedure) is then applied to derive flux estimates by inverse modelling. Consequently, only differences in the modelling of atmospheric transport may cause differences in the estimated fluxes. Here in our framework, we show that transport model errors lead to a discrepancy of 27 Tg yr -1 at the global scale, representing 5% of total methane emissions. At continental and annual scales, transport model errors are proportionally larger than at the global scale, with errors ranging from 36 Tg yr -1 in North America to 7 Tg yr -1 in Boreal Eurasia (from 23 to 48%, respectively). At the model grid-scale, the spread of inverse estimates can reach 150% of the prior flux. Therefore, transport model errors contribute significantly to overall uncertainties in emission estimates by inverse modelling, especially when small spatial scales are examined. Sensitivity tests have been carried out to estimate the impact of the measurement network and the advantage of higher horizontal resolution in transport models. The large differences found between methane flux estimates inferred in these different configurations highly question the consistency of transport model errors in current inverse systems.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1994

Goal Directed Model Inversion: Learning Within Domain Constraints

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome "would have been right if the outcome had been the desired one." The algorithm makes use of these intermediate "successes" to achieve the final goal. A unique and potentially very important feature of this algorithm is the ability to modify the output of the learning module to force upon it a desired syntactic structure. This differs from ordinary supervised learning in the following way: in supervised learning the exact desired output pattern must be provided. In GDMI instead, it is possible to require simply that the output obey certain rules, i.e., that it "make sense" in some way determined by the knowledge domain. The exact pattern that will achieve the desired outcome is then found by the system. The ability to impose rules while allowing the system to search for its own answers in the context of neural networks is potentially a major breakthrough in two ways: 1) it may allow the construction of networks that can incorporate immediately some important knowledge, i.e. would not need to learn everything from scratch as normally required at present, and 2) learning and searching would be limited to the areas where it is necessary, thus facilitating and speeding up the process. These points are illustrated with examples from robotic path planning and parametric design.

Colombano, Silvano P.↗
NASA NTRS1994

Goal Directed Model Inversion: Learning Within Domain Constraints

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome "would have been right if the outcome had been the desired one." The algorithm makes use of these intermediate "successes" to achieve the final goal. A unique and potentially very important feature of this algorithm is the ability to modify the output of the learning module to force upon it a desired syntactic structure. This differs from ordinary supervised learning in the following way: in supervised learning the exact desired output pattern must be provided. In GDMI instead, it is possible to require simply that the output obey certain rules, i.e., that it "make sense" in some way determined by the knowledge domain. The exact pattern that will achieve the desired outcome is then found by the system. The ability to impose rules while allowing the system to search for its own answers in the context of neural networks is potentially a major breakthrough in two ways: (1) it may allow the construction of networks that can incorporate immediately some important knowledge, i.e., would not need to learn everything from scratch as normally required at present; and (2) learning and searching would be limited to the areas where it is necessary, thus facilitating and speeding up the process. These points are illustrated with examples from robotic path planning and parametric design.

Colombano, Silvano P.↗
DOE OSTI2011

Top-down estimate of anthropogenic emission inventories and their interannual variability in Houston using a mesoscale inverse modeling technique

The 2000 and 2006 Texas Air Quality Study (TexAQS 2000 and 2006) field campaigns took place in eastern Texas in August-October of 2000 and 2006. Several flights of the National Oceanic and Atmospheric Administration (NOAA) and National Center for Atmospheric Research (NCAR) research aircraft were dedicated to characterizing anthropogenic emissions over Houston. Houston is known for having serious problems with non-attainment of air quality standards. We present a method that uses three models and aircraft observations to assess and improve existing emission inventories using an inverse modeling technique. We used 3-dimensional and 4-dimensional variational (3D-VAR and 4D-VAR) inverse modeling techniques based on a least-squares method to improve the spatial and temporal distribution of CO, NOy (sum of all reactive nitrogen compounds), and SO2 emissions predicted by the 4-km-resolution U.S. Environmental Protection Agency (EPA) National Emission Inventory (NEI) for 2005. Differences between the prior and posterior inventories are discussed in detail. We found that in 2006 the prior daytime emissions in the urban area of Houston have to be reduced by 40% {+-} 12% for CO and 7% {+-} 13% for NOy. Over the Houston Ship Channel, where industrial emissions are predominant, the prior emissions have to be reduced by 41% {+-} 15% for CO and 51% {+-} 9% for NOy. Major ports around Houston have their NOy emissions reduced as well, probably due to uncertainties in near-shore ship emissions in the EPA NEI inventory. Using the measurements from the two field campaigns, we assessed the interannual emission variability between 2000 and 2006. Daytime CO emissions from the Houston urban area have been reduced by 8% {+-} 20%, while the NOy emissions have increased by 20% {+-} 12% from 2000 to 2006. In the Houston Ship Channel, the daytime NOy emissions have increased by 13% {+-} 17%. Our results show qualitative consistencies with known changes in Houston emissions sources.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Generative Adversarial Networks and Mixture Density Networks-Based Inverse Modeling for Microstructural Materials Design

Abstract There are two broad modeling paradigms in scientific applications: forward and inverse. While forward modeling estimates the observations based on known causes, inverse modeling attempts to infer the causes given the observations. Inverse problems are usually more critical as well as difficult in scientific applications as they seek to explore the causes that cannot be directly observed. Inverse problems are used extensively in various scientific fields, such as geophysics, health care and materials science. Exploring the relationships from properties to microstructures is one of the inverse problems in material science. It is challenging to solve the microstructure discovery inverse problem, because it usually needs to learn a one-to-many nonlinear mapping. Given a target property, there are multiple different microstructures that exhibit the target property, and their discovery also requires significant computing time. Further, microstructure discovery becomes even more difficult because the dimension of properties (input) is much lower than that of microstructures (output). In this work, we propose a framework consisting of generative adversarial networks and mixture density networks for inverse modeling of structure–property linkages in materials, i.e., microstructure discovery for a given property. The results demonstrate that compared to baseline methods, the proposed framework can overcome the above-mentioned challenges and discover multiple promising solutions in an efficient manner.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Linear functional minimization for inverse modeling

In this paper, we present a novel inverse modeling strategy to estimate spatially distributed parameters of nonlinear models. The maximum a posteriori (MAP) estimators of these parameters are based on a likelihood functional, which contains spatially discrete measurements of the system parameters and spatiotemporally discrete measurements of the transient system states. The piecewise continuity prior for the parameters is expressed via Total Variation (TV) regularization. The MAP estimator is computed by minimizing a nonquadratic objective equipped with the TV operator. We apply this inversion algorithm to estimate hydraulic conductivity of a synthetic confined aquifer from measurements of conductivity and hydraulic head. The synthetic conductivity field is composed of a low-conductivity heterogeneous intrusion into a high-conductivity heterogeneous medium. Our algorithm accurately reconstructs the location, orientation, and extent of the intrusion from the steady-state data only. Finally, addition of transient measurements of hydraulic head improves the parameter estimation, accurately reconstructing the conductivity field in the vicinity of observation locations.

47 OTHER INSTRUMENTATION↗
NASA NTRS1994

Goal Directed Model Inversion: A Study of Dynamic Behavior

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome 0 "would have been right if the outcome had been the desired one." The algorithm then proceeds as follows: (1) store the action that produced the wrong outcome as a "target" (2) redefine the wrong outcome as a desired goal (3) submit the new desired goal to the system (4) compare the new action with the target action and modify the system by using a suitable algorithm for credit assignment (Back propagation in our example) (5) resubmit the original goal. Prior publications by our group in this area focused on demonstrating empirical results based on the inverse kinematic problem for a simulated robotic arm. In this paper we apply the inversion process to much simpler analytic functions in order to elucidate the dynamic behavior of the system and to determine the sensitivity of the learning process to various parameters. This understanding will be necessary for the acceptance of GDMI as a practical tool.

Colombano, Silvano P.↗
DOE OSTI2016

Evaluation of wetland methane emissions across North America using atmospheric data and inverse modeling

Existing estimates of methane (CH 4 ) fluxes from North American wetlands vary widely in both magnitude and distribution. In light of these differences, this study uses atmospheric CH 4 observations from the US and Canada to analyze seven different bottom-up, wetland CH 4 estimates reported in a recent model comparison project. We first use synthetic data to explore whether wetland CH 4 fluxes are detectable at atmospheric observation sites. We find that the observation network can detect aggregate wetland fluxes from both eastern and western Canada but generally not from the US. Based upon these results, we then use real data and inverse modeling results to analyze the magnitude, seasonality, and spatial distribution of each model estimate. The magnitude of Canadian fluxes in many models is larger than indicated by atmospheric observations. Many models predict a seasonality that is narrower than implied by inverse modeling results, possibly indicating an oversensitivity to air or soil temperatures. The LPJ-Bern and SDGVM models have a geographic distribution that is most consistent with atmospheric observations, depending upon the region and season. Lastly, these models utilize land cover maps or dynamic modeling to estimate wetland coverage while most other models rely primarily on remote sensing inundation data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2018

Estimating the reaction parameters of oil shale pyrolysis and oil shale grade using temperature transient analysis and inverse modeling

Grade of oil shale and reaction parameters of in-situ pyrolysis must be identified for the prediction of productivity before actual heating and production. Identification of oil shale grade and reaction parameters depends on laboratory experiments on core samples. However, laboratory-determined parameters can be different from those representing in-situ reservoir conditions. We use inverse modeling to determine oil shale grade and reaction parameters. The inversions are based on a forward model that simulates heat injection into a well. Temperature at the heating well is affected by a thermal skin effect as a result of a decrease of composite thermal conductivity around the heater due to the decomposition-induced porosity increase. Synthetic observations of heater temperature are generated from a forward simulation. Temperature difference and its derivative are used in synthetic inversions to estimate oil shale grade and parameters of active decomposition reactions with an error below 1%. Finally, the proposed methodology of inverse modeling is expected to successfully estimate the oil shale grade and reaction parameters without core sampling and subsequent surface experiments.

04 OIL SHALES AND TAR SANDS↗
DOE OSTI2016

Inverse modeling of ground surface uplift and pressure with iTOUGH-PEST and TOUGH-FLAC: The case of CO2 injection at In Salah, Algeria

Ground deformation, commonly seen in storage projects, carries useful information about processes occurring in the injection formation. The Krechba gas field at In Salah (Algeria) is one of the best-known sites for studying ground surface deformation during geological carbon storage. At this first industrial-scale on-shore CO 2 demonstration project, satellite-based ground-deformation monitoring data of high quality are available and used to study the large-scale hydrological and geomechanical response of the system to injection. In this work, we carry out coupled fluid flow and geomechanical simulations to understand the uplift at three different CO 2 injection wells (KB-501, KB-502, KB-503). Previous numerical studies focused on the KB-502 injection well, where a double-lobe uplift pattern has been observed in the ground-deformation data. The observed uplift patterns at KB-501 and KB-503 have single-lobe patterns, but they can also indicate a deep fracture zone mechanical response to the injection.The current study improves the previous modeling approach by introducing an injection reservoir and a fracture zone, both responding to a Mohr-Coulomb failure criterion. In addition, we model a stress-dependent permeability and bulk modulus, according to a dual continuum model. Mechanical and hydraulic properties are determined through inverse modeling by matching the simulated spatial and temporal evolution of uplift to InSAR observations as well as by matching simulated and measured pressures. The numerical simulations are in agreement with both spatial and temporal observations. The estimated values for the parameterized mechanical and hydraulic properties are in good agreement with previous numerical results. In addition, the formal joint inversion of hydrogeological and geomechanical data provides measures of the estimation uncertainty.

58 GEOSCIENCES↗
DOE OSTI2019

A heuristic model inversion for coupled thermo-hydro-mechanical modelling of triaxial experiments

As multiphysics geomechanical models get developed, their increasing complexity and number of parameters make it particularly difficult to calibrate against experimental data. In this contribution, we present a heuristic workflow to invert for parameters of a coupled Thermo-Hydro-Mechanical (THM) model in a way that helps the theoretical modellers refine their definition of the underlying elasto-visco-plastic model itself. We apply this workflow to the calibration of deviatoric and volumetric data for two sets of triaxial experiments on mudstone and sandstone. Here, we show that beyond the calibration of well-defined parameters of the THM model, one or two scaling factors of the experimentally obtained yield surface and two more factors to capture the confinement and pore pressure dependency of the flow law lead to the satisfactory matching of series of experiments at varying confinements for two different rock types. Using this physical model, we also show that tracking the volumetric component of the mechanical power in the numerical simulations might allow reducing the number of experiments required to calibrate the model.

58 GEOSCIENCES↗
NASA NTRS1992

Shinnery oak bidirectional reflectance properties and canopy model inversion

Field measurements are presented, together with the results of a 3D canopy-model inversion for sand shinnery oak community in western Texas. The spectral bidirectional radiance measurements were in three spectral channels encompassing both the complete land surface and sky hemispheres. The changes in canopy reflectance that occur with variations in solar zenith angle and view direction for two seasons of the year were evaluated, and the 3D radiation-interaction model was inverted to estimate the oak leaf area index and canopy density from the reflectance data.

Deering, Donald W.↗
NASA NTRS1990

Aircraft automatic flight control system with model inversion

A simulator study was conducted to verify the advantages of a Newton-Raphson model-inversion technique as a design basis for an automatic trajectory control system in an aircraft with highly nonlinear characteristics. The simulation employed a detailed mathematical model of the aerodynamic and propulsion system performance characteristics of a vertical-attitude takeoff and landing tactical aircraft. The results obtained confirm satisfactory control system performance over a large portion of the flight envelope. System response to wind gusts was satisfactory for various plausible combinations of wind magnitude and direction.

Smith, G. A.↗