Search NASA⌕ Search

SEARCH · Search NASA

Results for “Model error”

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 1,279 records · Page 71

IMPROVED SPECTROPHOTOMETRIC CALIBRATION OF THE SDSS-III BOSS QUASAR SAMPLE

We present a model for spectrophotometric calibration errors in observations of quasars from the third generation of the Sloan Digital Sky Survey Baryon Oscillation Spectroscopic Survey (BOSS) and describe the correction procedure we have developed and applied to this sample. Calibration errors are primarily due to atmospheric differential refraction and guiding offsets during each exposure. The corrections potentially reduce the systematics for any studies of BOSS quasars, including the measurement of baryon acoustic oscillations using the Ly α forest. Our model suggests that, on average, the observed quasar flux in BOSS is overestimated by ∼19% at 3600 Å and underestimated by ∼24% at 10,000 Å. Our corrections for the entire BOSS quasar sample are publicly available.

79 ASTRONOMY AND ASTROPHYSICS↗

Probabilistic Wind Power Ramp Forecasting Based on a Scenario Generation Method: Preprint

Wind power ramps (WPRs) are particularly important in the management and dispatch of wind power, and they are currently drawing the attention of balancing authorities. With the aim to reduce the impact of WPRs for power system operations, this paper develops a probabilistic ramp forecasting method based on a large number of simulated scenarios. An ensemble machine learning technique is first adopted to forecast the basic wind power forecasting scenario and calculate the historical forecasting errors. A continuous Gaussian mixture model (GMM) is used to fit the probability distribution function (PDF) of forecasting errors. The cumulative distribution function (CDF) is analytically deduced. The inverse transform method based on Monte Carlo sampling and the CDF is used to generate a massive number of forecasting error scenarios. An optimized swinging door algorithm is adopted to extract all the WPRs from the complete set of wind power forecasting scenarios. The probabilistic forecasting results of ramp duration and start time are generated based on all scenarios. Numerical simulations on publicly available wind power data show that within a predefined tolerance level, the developed probabilistic wind power ramp forecasting method is able to predict WPRs with a high level of sharpness and accuracy.

17 WIND ENERGY↗

Probabilistic Wind Power Ramp Forecasting Based on a Scenario Generation Method

Wind power ramps (WPRs) are particularly important in the management and dispatch of wind power and currently drawing the attention of balancing authorities. With the aim to reduce the impact of WPRs for power system operations, this paper develops a probabilistic ramp forecasting method based on a large number of simulated scenarios. An ensemble machine learning technique is first adopted to forecast the basic wind power forecasting scenario and calculate the historical forecasting errors. A continuous Gaussian mixture model (GMM) is used to fit the probability distribution function (PDF) of forecasting errors. The cumulative distribution function (CDF) is analytically deduced. The inverse transform method based on Monte Carlo sampling and the CDF is used to generate a massive number of forecasting error scenarios. An optimized swinging door algorithm is adopted to extract all the WPRs from the complete set of wind power forecasting scenarios. The probabilistic forecasting results of ramp duration and start-time are generated based on all scenarios. Numerical simulations on publicly available wind power data show that within a predefined tolerance level, the developed probabilistic wind power ramp forecasting method is able to predict WPRs with a high level of sharpness and accuracy.

17 WIND ENERGY↗

Comparison of Measured Direct Normal Irradiance Data to Calculated Using the Disc Model for the Texas Panhandle

It is important that we convert from fossil energy to wind and solar energy. Wind energy is currently the cheapest form of renewable energy, but it is usually a poor match to utility electrical generation in the Great Plains of the United States. Concentrating solar power (CSP) plants are a better match to utility electrical loading, but so far none of these plants have been constructed in the Great Plains. Part of the reason no CSP plants have been constructed is lack of confidence in the solar resource data required (direct normal irradiance or DNI for short). There are several models for converting global solar radiation to DNI, and one of them is the DISC model. Both DNI data and global radiation data were collected at the USDA-Agricultural Research Laboratory at Bushland, Texas, for the years 2009 and 2010. Using the DISC model and measured global radiation data, DNI was calculated and compared to measured DNI data. The total annual solar energy estimated with the model was 5% greater than what was actually measured which was fairly close. However, on a monthly or hourly basis, the percentage difference could be as high as 25%. Analysis of individual months and days resulted in some recommendations for improving the DISC model. Some of the poor correlations between measured solar energy and calculated solar energy using the DISC model implied that another model (DIRINDEX) might be better. Therefore, additional work is planned to investigate the DIRINDEX model with the same data set to see if the correlation improves. This work may potentially lead to the increased use of solar energy for electrical generation, which will decrease the use of fossil fuels and the environmental pollution from fossil fuel power plants. Direct normal irradiance (DNI) is required in the performance estimation of concentrating solar energy systems. The objective of this paper is to compare measured and modeled DNI data for a site in the Texas Panhandle (Bushland, Texas) to determine the accuracy of the model and where improvements might be required. The DNI model used was the Direct Insolation Simulation Code (DISC) which was developed at the National Renewable Energy Lab (NREL) in Boulder, Colorado. We have determined that a portion of the DISC standard deviation error is due to inaccurate modeling of the first and last hour of the day, but other inaccuracies of the DISC model are also discussed. The accuracy of the DISC model improves as the time scale increases (e.g. hour to month to year) due to errors cancelling, and for Bushland, Texas, the DISC model during 2009 predicted 5% more solar energy than the measured DNI.

14 SOLAR ENERGY↗

Sci—Sat AM: Stereo — 09: Accuracy of Liver Cancer Treatment on Cyberknife® with Synchrony™ Optical Tracking Throughout the Respiratory Cycle

The Cyberknife® robotic stereotactic body radiation therapy system is well-suited for treating liver lesions over the respiratory cycle as it includes room-mounted orthogonal x-ray tracking of internal fiducial markers and optical tracking of external markers. The Synchrony™ software generates a model of internal target positions during patient respiration and correlates it to the external optical tracking system for real-time optical-based position corrections of the linear accelerator during beam delivery. Although clinical studies have provided preliminary outcomes for liver lesions treated with the Cyberknife system, to date, there is little data demonstrating the ability of the Synchrony software to track targets in the liver, which deforms throughout the respiratory cycle. In this study, we investigated the respiratory motion model performance for predicting tumour motion. We conducted a retrospective analysis of fifteen liver cancer patients treated on the Cyberknife using the Synchrony optical tracking system. We analyzed Cyberknife tracking information stored in the log files to extract the left-right (LR), anterior-posterior (AP) and superior-inferior (SI) correlation errors between the model-predicted position and the internal fiducial centroid position determined by x-ray imaging. Only translational tracking and corrections were applied during treatment. Overall, the correlation errors were greatest in the SI direction. We calculated radial correlation errors, and determined that the 95{sup th}, 98{sup th} and 99{sup th} percentile errors were 3.4 mm, 4.4 mm and 5.1 mm, respectively. Based on translational correlation tracking errors we expect the clinical target volume will be within 3.4 mm of the planning target volume for 95 % of beam delivery time.

60 APPLIED LIFE SCIENCES↗

Comments on “Comparison of energy consumption in desalination by capacitive deionization and reverse osmosis”

Capacitive deionization (CDI) is an emerging brackish water desalination technology in which ions are removed from water by electrostatically adsorbing them onto porous electrodes. Recently, Qin et al. presented an estimate of energy consumption for CDI schemes, and compared the energy consumption of CDI to the energy consumption of a particular reverse osmosis operation scheme which they propose. Qin et al. estimate very high values of energy consumption in CDI for brackish water desalination (e.g. ~1 to 2 g/L salt solution), and these values and trends do not agree with published experiments for CDI or with experimentally validated models for CDI. The current comment identifies important scaling errors and incorrect values in their model resistance parameters, and then discusses unphysical trends predicted by their model. We do not attempt provide a comprehensive review of their model, operation, estimates, and figures, but rather concentrate on several key issues. We present our comments in two sections dealing with resistances and energy metrics, respectively.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Structure Sensitive Reaction Kinetics of Chiral Molecules on Intrinsically Chiral Surfaces

Enantiospecific heterogeneous catalysis utilizes chiral surfaces to resolve enantiomers via structure sensitive surface chemistry. The catalyst design challenge is the identification of chiral surface structures that maximize enantiospecificity. Herein, we develop data driven models for the enantiospecificity of tartaric acid reactions on chiral Cu(hkl) R&S surfaces. Measurements of enantiospecific rate constants were obtained by using curved Cu(hkl) R&S surfaces that enable kinetic measurements on hundreds of chiral surface orientations. One model uses feature vectors derived from generalized coordination numbers to capture the local structure around Cu atoms exposed by the Cu(hkl) R&S surfaces. The second model introduces the use of chiral cubic harmonic functions to capture the symmetry constraints of the face-centered cubic Cu structure. The model using 58 generalized coordination numbers has a fitting error similar to that of the model using only 5 cubic harmonic functions. The two models predict maxima in the enantiospecificity on surfaces with very similar surface orientations. The models developed in this work are applicable for any enantiospecific reaction happening on any chiral material with a cubic lattice structure, opening the way to understanding the surface structure sensitivity of the enantiospecific reaction kinetics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Reduced‐Order Modeling for Linearized Representations of Microphysical Process Rates

Abstract Representing cloud microphysical processes in large scale atmospheric models is challenging because many processes depend on the details of the droplet size distribution (DSD, the spectrum of droplets with different sizes in a cloud). While full or partial statistical moments of droplet size distributions are the typical variables used in bulk models, prognostic moments are limited in their ability to represent microphysical processes across the range of conditions experienced in the atmosphere. Microphysical parameterizations employing prognostic moments are known to suffer from structural uncertainty in their representations of inherently higher dimensional cloud processes, which limit model fidelity and lead to forecasting errors. Here we investigate how data‐driven reduced‐order modeling can be used to learn predictors for microphysical process rates in bulk microphysics schemes in an unsupervised manner from higher dimensional bin distributions. Using simulations characteristic of marine stratiform clouds, we simultaneously learn lower dimensional representations of droplet size distributions and predict the evolution of the microphysical state of the system. Droplet collision‐coalescence, the main process for generating warm rain, is estimated to have an intrinsic dimension of three. This intrinsic dimension provides a lower limit on the number of degrees of freedom needed to accurately represent collision‐coalescence in models. We demonstrate how deep learning based reduced‐order modeling can be used to discover intrinsic coordinates describing the microphysical state of the system, where process rates such as collision‐coalescence are globally linearized. These implicitly learned representations of the DSD retain more information about the DSD than typical moment‐based representations.

54 ENVIRONMENTAL SCIENCES↗

A machine learning approach for efficient multi-dimensional integration

Many physics problems involve integration in multi-dimensional space whose analytic solution is not available. The integrals can be evaluated using numerical integration methods, but it requires a large computational cost in some cases, so an efficient algorithm plays an important role in solving the physics problems. We propose a novel numerical multi-dimensional integration algorithm using machine learning (ML). After training a ML regression model to mimic a target integrand, the regression model is used to evaluate an approximation of the integral. Then, the difference between the approximation and the true answer is calculated to correct the bias in the approximation of the integral induced by ML prediction errors. Because of the bias correction, the final estimate of the integral is unbiased and has a statistically correct error estimation. Three ML models of multi-layer perceptron, gradient boosting decision tree, and Gaussian process regression algorithms are investigated. The performance of the proposed algorithm is demonstrated on six different families of integrands that typically appear in physics problems at various dimensions and integrand difficulties. The results show that, for the same total number of integrand evaluations, the new algorithm provides integral estimates with more than an order of magnitude smaller uncertainties than those of the VEGAS algorithm in most of the test cases.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Hierarchical top-down bottom-up calibration with consideration for uncertainty and inter-scale discrepancy of Peierls stress of bcc Fe

A hierarchical multiscale model of plasticity in single crystal bcc Fe, framed at the mesoscale pertaining to a statistically representative set of dislocations for each slip system, relies on specification of a set of parameters which are informed using both bottom-up (BU) (atomistic) and top-down (TD) (experimental) information. We term this specification process a 'connection' between BU and TD pathways to inform the mesoscale model. The connection is considered in the presence of error, uncertainty, and discrepancy between the models. Furthermore, we expand upon a previously developed reconciled TD and BU calibration method to account for anticipated discrepancy between information from different length scales. The results of a previously formulated likelihood-based connection test of the multiscale model suggest a 'missing link' may be responsible for part of the inter-scale discrepancy. In this case, this link is assumed to be a relation between the unit-process (single dislocation line) of coordinated kink-pair nucleation on a screw dislocation segment and the many-body dislocation process which manifests in the onset of experimentally observed plastic deformation in a single crystal. A physics-informed discrepancy layer is formulated to improve the connection in the presence of additional persistent uncertainties. This physics-informed hypothesis testing is demonstrated as an alternative to fully data-driven search methods, particularly applicable to datasets of limited size that are often encountered in such multiscale material modeling applications. Discrete dislocation dynamics simulations are used to investigate the effect of the Peierls stress on straight screw dislocations and in the operation of Frank–Read sources. Discussion concerns the critical role of uncertainty quantification in providing a basis for this form of incremental hypothesis testing.

42 ENGINEERING↗

Least H 2 norm updating of quadratic interpolation models for derivative-free trust-region algorithms

One particular class of derivative-free optimization algorithms is trust-region algorithms based on quadratic models given by the under-determined interpolation. Different techniques in updating the quadratic model from iteration to iteration will give different interpolation models. We propose a new way to update the quadratic model by minimizing the $H^{2}$ norm of the difference between neighboring quadratic models. The motivation for applying the $H^{2}$ norm is given. The theoretical properties of our new updating technique are also presented. We propose the projection in the sense of $H^{2}$ norm and the interpolation error analysis of our model function. We obtain the coefficients of the quadratic model function using the Karush–Kuhn–Tucker (KKT) conditions. Numerical results show the advantages of our model on the test set considered, and the derivative-free algorithms based on our least $H^{2}$ norm updating quadratic model functions can solve test problems with fewer function evaluations than the algorithm based on the least Frobenius norm updating model and the other compared methods.

derivative-free optimization↗

A Method for Estimating Time-Series PV Production Loss From Solar Tracking Failures

Solar tracking system failures can dramatically reduce photovoltaic (PV) system energy output through suboptimal irradiance capture and self-shading loss. Detecting tracker failures is not always straightforward; unlike other common PV system failures like inverter outages and blown fuses which take portions of a system completely offline, tracker failures only partially reduce output power and can go unnoticed as a result. Additionally, tracker failures can take one of several forms, each with their own loss characteristics. Here we present two methods of detecting tracker failure events from time-series production data and a method for estimating the associated production loss. Compared with existing detection methods, the proposed power-based detection method showed 0-17% reductions in Type I error rate depending on weather conditions. Here, the loss model estimated production loss with low error (mean bias error = -2.3 %, root-mean-squared error = 6%) in a rudimentary validation.

14 SOLAR ENERGY↗

An error estimation driven adaptive tetrahedral workflow for full engineering models

Tetrahedral finite element workflows have the potential to drastically reduce time to solution for computational solid mechanics simulations when compared to traditional hexahedral finite element analogues. A recently developed, higher-order composite tetrahedral element has shown promise in the space of incompressible computational plasticity. Mesh adaptivity has the potential to increase solution accuracy and increase solution robustness. In this work, we demonstrate an initial strategy to perform conformal mesh adaptivity for this higher-order composite tetrahedral element using well-established mesh modification operations for linear tetrahedra. We propose potential extensions to improve this initial strategy in terms of robustness and accuracy.

42 ENGINEERING↗

Application of New Los Alamos OPLIB Opacities in Solar Modeling Using the Mesa Code

The Sun provides the only local laboratory to test our understanding of stellar physics. The production, then, of valid dynamical models of the Sun is critical to how we think about other stars. Despite improvements in modeling methodologies and abundance measurements, solar models continue to reproduce errors in convection zone depth and surface helium abundance compared to helioseismic observations. Using the 1D stellar evolution code MESAstar, this work first evaluates the effect of changes to timestep resolution and mixing length theory on model outputs. A standard test case is then constructed with both GS98 and AGSS09 abundances using OP and OPAL opacities before implementing the new Los Alamos OPLIB opacities. It is shown that, when using GS98 abundances, OPLIB opacities improve agreement with helioseismic inference of the convection zone base radius, R czb , by 1.68-σ and 1.19-σ compared to OP and OPAL, respectively, while worsening agreement with the inferred envelope helium mass fraction, Y surf , by 0.63-σ and 0.98-σ for OP and OPAL.

14 SOLAR ENERGY↗

THE UBV(RI){sub C} COLORS OF THE SUN

Photometric data in the UBV(RI){sub C} system have been acquired for 80 solar analog stars for which we have previously derived highly precise atmospheric parameters T{sub eff}, log g, and [Fe/H] using high-resolution, high signal-to-noise ratio spectra. UBV and (RI){sub C} data for 46 and 76 of these stars, respectively, are published for the first time. Combining our data with those from the literature, colors in the UBV(RI){sub C} system, with {approx_equal} 0.01 mag precision, are now available for 112 solar analogs. Multiple linear regression is used to derive the solar colors from these photometric data and the spectroscopically derived T{sub eff}, log g, and [Fe/H] values. To minimize the impact of systematic errors in the model-dependent atmospheric parameters, we use only the data for the 10 stars that most closely resemble our Sun, i.e., the solar twins, and derive the following solar colors: (B - V){sub Sun} = 0.653 {+-} 0.005, (U - B){sub Sun} = 0.166 {+-} 0.022, (V - R){sub Sun} = 0.352 {+-} 0.007, and (V - I){sub Sun} = 0.702 {+-} 0.010. These colors are consistent, within the 1{sigma} errors, with those derived using the entire sample of 112 solar analogs. We also derive the solar colors using the relation between spectral-line-depth ratios and observed stellar colors, i.e., with a completely model-independent approach, and without restricting the analysis to solar twins. We find (B - V){sub Sun} = 0.653 {+-} 0.003, (U - B){sub Sun} = 0.158 {+-} 0.009, (V - R){sub Sun} = 0.356 {+-} 0.003, and (V - I){sub Sun} = 0.701 {+-} 0.003, in excellent agreement with the model-dependent analysis.

79 ASTRONOMY AND ASTROPHYSICS↗

Environmental drivers of methane fluxes from an urban temperate wetland park

Abstract Methane (CH 4 ) emissions were measured at the Wilma H. Schiermeier Olentangy River Wetland Research Park (ORWRP) over three summers and two winters using an eddy covariance system. We used an empirical model to determine the main environmental drivers of methane emissions. Methane emissions covary strongly with water vapor fluxes, CO 2 fluxes, and soil temperature. We adjust our models to account for the heterogeneous environment of the wetland by including the flux footprint distribution among different microsites as a predictive variable in the methane model. We used a forward linear stepwise model in combination with an Akaike information criteria‐based model selection process and neural network modeling to determine which environmental variables are most effective in modeling methane emissions in our site. Different models and environmental variables best represented methane fluxes in the winter and summer and also during the day or night within each season. We parameterized an optimal empirical model for methane emissions from the ORWRP that is used for gap filling of site‐level methane fluxes over 2 years. Some of the most effective variables for modeling methane were carbon, water vapor, and heat fluxes, all of which typically have the same data gaps as the time series of methane flux. In order to determine if these variables were useful for modeling methane despite the additional gap‐filling error, we determined through an error propagation experiment that eddy covariance gap‐filling models for methane may be best developed by including other gap‐filled fluxes as predictors, despite the high level of shared gaps and subsequent gap‐fill error propagation.

Morin, T. H.↗

Error field impact on mode locking and divertor heat flux in NSTX-U

During the 2016 NSTX-U experimental campaign, locked modes in the plasma edge presented clear evidence of the presence of error fields. Extensive metrology and plasma response modeling with IPEC and M3D-C1 have been conducted to understand the various sources of error fields in NSTX-U as built in 2016, and to determine which of these sources have the greatest effect on the plasma. In particular, modeling finds that the error field from misalignment of the toroidal field (TF) coils may have a significant effect on the plasma. The response to the TF error field is shown to depend on the presence of a q = 1 surface, in qualitative agreement with experimental observations. It is found that certain characteristics of the TF error field present new challenges for error field correction. Specifically, the error field spectrum differs significantly from that of coils on the low-field side (such as the NSTX-U error field correction coils), and does not resonate strongly with the dominant kink mode, thus potentially requiring a multi-mode correction. Moreover, to mitigate heat fluxes using poloidal flux expansion, the pitch angle at the divertor plates must be small (~1°). It is shown that uncorrected error fields may result in potentially significant local perturbation to the pitch angle. Estimates for coil alignment tolerances in NSTX-U are derived based on consideration of both heat flux and core resonant fields independently.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Predictive RANS simulations via Bayesian Model-Scenario Averaging

The turbulence closure model is the dominant source of error in most Reynolds-Averaged Navier–Stokes simulations, yet no reliable estimators for this error component currently exist. Here we develop a stochastic, a posteriori error estimate, calibrated to specific classes of flow. It is based on variability in model closure coefficients across multiple flow scenarios, for multiple closure models. The variability is estimated using Bayesian calibration against experimental data for each scenario, and Bayesian Model-Scenario Averaging (BMSA) is used to collate the resulting posteriors, to obtain a stochastic estimate of a Quantity of Interest (QoI) in an unmeasured (prediction) scenario. The scenario probabilities in BMSA are chosen using a sensor which automatically weights those scenarios in the calibration set which are similar to the prediction scenario. The methodology is applied to the class of turbulent boundary-layers subject to various pressure gradients. For all considered prediction scenarios the standard-deviation of the stochastic estimate is consistent with the measurement ground truth. Furthermore, the mean of the estimate is more consistently accurate than the individual model predictions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗