Search NASA⌕ Search

SEARCH · Search NASA

Results for “computable error estimate”

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

Combined High Spectral Resolution Lidar and Millimeter Wavelength Radar Measurement of Ice Crystal Precipitation

The goal of this research has been to improve measurements of snowfall using a combination of millimeter-wavelength radar and High Spectral Resolution Lidar (HSRL) Observations. Snowflakes are large compared to the 532nm HSRL wavelength and small compared to the 3.2 and 8.6 mm wavelength radars used in this study. This places the particles in the optical scattering regime of the HSRL, where extinction cross-section is proportional to the projected area of the particles, and in the Rayleigh regime for the radar, where the backscatter cross-section is proportional to the mass-squared of the particles. Forming a ratio of the radar measured cross-section to the HSRL measured cross section eliminates any dependence on the number of scattering particles, yielding a quantity proportional to the average mass-squared of the snowflakes over the average area of the flakes. Using simultaneous radar measurements of particle fall velocities, which are dependent particle mass and cross-sectional area it is possible to derive the average mass of the snow flakes, and with the radar measured fall velocities compute the snowfall rate. Since this retrieval requires the optical extinction cross-section we began by considering errors this quantity. The HSRL is particularly good at measuring the backscatter cross-section. In previous studies of snowfall in the high Arctic were able to estimate the extinction cross-section directly as a fixed ratio to the backscatter cross-section. Measurements acquired in the STORMVEX experiment in Colorado showed that this approach was not valid in mid-latitude snowfalls and that direct measurement of the extinction cross-section is required. Attempts to measure the extinction directly uncovered shortcomings in thermal regulation and mechanical stability of the newly deployed DOE HSRL systems. These problems were largely mitigated by modifications installed in both of the DOE systems. We also investigated other sources of error in the HSRL direct measurement of extinction (see appendix II of this report). We also developed improved algorithms to extract extinction from the HSRL data. These have been installed in the standard HSRL data processing software and are now available to all users of HSRL data. Validation of snowfall measurements has proven difficult due to the unreliability of conventional snowfall measurements coupled with the complexity of considering the vast variety of snowflake geometries. It was difficult to tell how well the algorithm’s approach to accommodating differences in snowflakes was working without good measurements for comparison. As a result, we decided to apply this approach to the somewhat simpler, but scientifically important, problem of drizzle measurement. Here the particle shape is known and the conventional measurement are more reliable. These algorithms where successfully applied to drizzle data acquired during the ARM MAGIC study of marine stratus clouds between California and Hawaii (see Appendix I). This technique is likely to become a powerful tool for studying lifetime of the climatically important marine stratus clouds.

47 OTHER INSTRUMENTATION↗

Notes on Regression Analysis for Radar Parameter Estimation

A fundamental task of radar, beyond merely detecting a target, is to estimate some parameters associated with it. For example, this might include range, direction, velocity, etc. In any case, multiple measurements, often noisy, need to be processed to yield a ‘best estimate’ of the parameter. A common mathematical method for doing so is called “Regression” analysis. The goal is to minimize the expected squared error in the estimate. Even when alternate algorithms are considered, the least squared-error regression analysis is the benchmark against which alternatives are compared.

47 OTHER INSTRUMENTATION↗

An analysis of ridge axis gravity roughness and spreading rate

Fast and slow spreading ridges have radically different morphologic and gravimetric characteristics. In this study, altimeter measurements from the Geosat Exact Repeat Mission are used to investigate spreading rate dependence of the ridge axis gravity field. Gravity roughness provides an estimate of the amplitude of the gravity anomaly and is robust to small errors in the location of the ridge axis. Gravity roughness as a weighted root mean square of the vertical deflection at 438 ridge crossings on the mid-ocean ridge system is computed. Ridge axis gravity anomalies show a decrease in amplitude with increasing spreading rate up to an intermediate rate of about 60-80 mm/yr and almost no change at higher rates; overall the roughness decreases by a factor of 10 between the lowest and highest rates. In addition to the amplitude decrease, the range of roughness values observed at a given spreading rate shows a similar order of magnitude decrease with transition between 60 and 80 mm/yr. The transition of ridge axis gravity is most apparent at three relatively unexplored locations on the Southeast Indian Ridge and the Pacific-Antarctic Rise; on these intermediate rate ridges the transition occurs abruptly across transform faults.

Small, Christopher↗

A parameterization for longwave surface radiation from satellite data - Recent improvements

Several improvements have been made recently to the parameterization for surface longwave radiation described by Gupta (1989). Model constants have been modified in order to use meteorological data from the International Satellite Cloud Climatology Project instead of from the TIROS Operational Vertical Sounder data, primarily to take advantage of the vastly superior cloud information available from the former. Additional modifications were made to improve the estimation of cloud effect in the presence of low-level clouds. The latter modifications reduced the systematic error of the overcast-sky fluxes from 10.0 to 1.7 W/sq m and the random error from +/- 18.9 to +/- 6.3 W/sq m when compared to the fluxes computed with a detailed radiative transfer model.

Gupta, Shashi K.↗

A Nonlinear Adaptive Filter for Gyro Thermal Bias Error Cancellation

Deterministic errors in angular rate gyros, such as thermal biases, can have a significant impact on spacecraft attitude knowledge. In particular, thermal biases are often the dominant error source in MEMS gyros after calibration. Filters, such as J\,fEKFs, are commonly used to mitigate the impact of gyro errors and gyro noise on spacecraft closed loop pointing accuracy, but often have difficulty in rapidly changing thermal environments and can be computationally expensive. In this report an existing nonlinear adaptive filter is used as the basis for a new nonlinear adaptive filter designed to estimate and cancel thermal bias effects. A description of the filter is presented along with an implementation suitable for discrete-time applications. A simulation analysis demonstrates the performance of the filter in the presence of noisy measurements and provides a comparison with existing techniques.

Galante, Joseph M.↗

Dynamic Adjustment of Simulation Parameters for Efficient Vehicle Prognostics

Model-based diagnostics and prognostics rely on state estimation and uncertainty management algorithms to produce useful information for system operators and maintainers. This information enables more informed operational decisions, condition-based maintenance, and overall mission safety assurance. Typically, uncertainty is associated with vehicle state-of-health estimation and prediction results because of modeling errors, internal or external sources of noise, and sensor inaccuracy. Probabilistic uncertainty management methods including Sequential Monte Carlo simulation are commonly used to reason about state-of-health estimates and predictions in the presence of these sources of uncertainty. However, such algorithms can be computationally expensive as they require a very large number of samples to obtain a sufficiently accurate quantification of the end of life probability distribution. As a result, highly mobile autonomous systems that leverage the prognostic results for mission-level replanning are often constrained in their processing capability because of these computationally expensive simulation approaches. Therefore, in this paper, we investigate algorithmic methods for dynamically adjusting simulation time step as well as number of samples to achieve highly efficient prognostic results while maintaining results accuracy. Results obtained from simulated flight experiments of an electric unmanned aerial vehicle are presented to verify the efficacy of such algorithms.

Prognostics↗

Incorporation of Orbital Dynamics to Improve Wide-Area Differential GPS

A powerful technique to compute precise GPS satellite orbits for the FAA Wide-Area Augmentation System (WAAS) has been evaluated. This dynamical technique estimates GPS satellite states from a long history of measurements not only yields accurate orbit solutions, but it enables separation of orbit and satellite clock errors and yields orbit accuracies of better than one meter within the service volume, as compared to the JPL precise ephemerides.

Orbital↗

Fully automated calibration for acoustic pyrometry

A method of calibrating transceiver positions inside an acoustic pyrometry measuring vessel that contains a plurality of transceivers, includes determining (40) a speed of sound in the acoustic pyrometry measuring vessel from a temperature and gas composition of a gas inside the acoustic pyrometry measuring vessel, acquiring (41) time-of-flight (TOFs) Δti,j measurements from a plurality of pairs i,j of transceivers inside the acoustic pyrometry measuring vessel, estimating (42) a radius of the acoustic pyrometry measuring vessel from an average of the acquired TOF measurements, and using (43) an estimated radius of the acoustic pyrometry measuring vessel to estimate errors Δθj of displacement angles of the transceivers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications

Abstract Robust quantification of predictive uncertainty is a critical addition needed for machine learning applied to weather and climate problems to improve the understanding of what is driving prediction sensitivity. Ensembles of machine learning models provide predictive uncertainty estimates in a conceptually simple way but require multiple models for training and prediction, increasing computational cost and latency. Parametric deep learning can estimate uncertainty with one model by predicting the parameters of a probability distribution but does not account for epistemic uncertainty. Evidential deep learning, a technique that extends parametric deep learning to higher-order distributions, can account for both aleatoric and epistemic uncertainties with one model. This study compares the uncertainty derived from evidential neural networks to that obtained from ensembles. Through applications of the classification of winter precipitation type and regression of surface-layer fluxes, we show evidential deep learning models attaining predictive accuracy rivaling standard methods while robustly quantifying both sources of uncertainty. We evaluate the uncertainty in terms of how well the predictions are calibrated and how well the uncertainty correlates with prediction error. Analyses of uncertainty in the context of the inputs reveal sensitivities to underlying meteorological processes, facilitating interpretation of the models. The conceptual simplicity, interpretability, and computational efficiency of evidential neural networks make them highly extensible, offering a promising approach for reliable and practical uncertainty quantification in Earth system science modeling. To encourage broader adoption of evidential deep learning, we have developed a new Python package, Machine Integration and Learning for Earth Systems (MILES) group Generalized Uncertainty for Earth System Science (GUESS) (MILES-GUESS) ( https://github.com/ai2es/miles-guess ), that enables users to train and evaluate both evidential and ensemble deep learning. Significance Statement This study demonstrates a new technique, evidential deep learning, for robust and computationally efficient uncertainty quantification in modeling the Earth system. The method integrates probabilistic principles into deep neural networks, enabling the estimation of both aleatoric uncertainty from noisy data and epistemic uncertainty from model limitations using a single model. Our analyses reveal how decomposing these uncertainties provides valuable insights into reliability, accuracy, and model shortcomings. We show that the approach can rival standard methods in classification and regression tasks within atmospheric science while offering practical advantages such as computational efficiency. With further advances, evidential networks have the potential to enhance risk assessment and decision-making across meteorology by improving uncertainty quantification, a longstanding challenge. This work establishes a strong foundation and motivation for the broader adoption of evidential learning, where properly quantifying uncertainties is critical yet lacking.

Schreck, John S.↗

Computing the partition function, ensemble averages, and density of states for lattice spin systems by sampling the mean

An algorithm to approximately calculate the partition function (and subsequently ensemble averages) and density of states of lattice spin systems through non-Monte-Carlo random sampling is developed. This algorithm (called the sampling-the-mean algorithm) can be applied to models where the up or down spins at lattice nodes interact to change the spin states of other lattice nodes, especially non-Ising-like models with long-range interactions such as the biological model considered here. Because it is based on the Central Limit Theorem of probability, the sampling-the-mean algorithm also gives estimates of the error in the partition function, ensemble averages, and density of states. Easily implemented parallelization strategies and error minimizing sampling strategies are discussed. The sampling-the-mean method works especially well for relatively small systems, systems with a density of energy states that contains sharp spikes or oscillations, or systems with little a priori knowledge of the density of states.

97 MATHEMATICS AND COMPUTING↗

Dynamic Method for Identifying Collected Sample Mass

G-Sample is designed for sample collection missions to identify the presence and quantity of sample material gathered by spacecraft equipped with end effectors. The software method uses a maximum-likelihood estimator to identify the collected sample's mass based on onboard force-sensor measurements, thruster firings, and a dynamics model of the spacecraft. This makes sample mass identification a computation rather than a process requiring additional hardware. Simulation examples of G-Sample are provided for spacecraft model configurations with a sample collection device mounted on the end of an extended boom. In the absence of thrust knowledge errors, the results indicate that G-Sample can identify the amount of collected sample mass to within 10 grams (with 95-percent confidence) by using a force sensor with a noise and quantization floor of 50 micrometers. These results hold even in the presence of realistic parametric uncertainty in actual spacecraft inertia, center-of-mass offset, and first flexibility modes. Thrust profile knowledge is shown to be a dominant sensitivity for G-Sample, entering in a nearly one-to-one relationship with the final mass estimation error. This means thrust profiles should be well characterized with onboard accelerometers prior to sample collection. An overall sample-mass estimation error budget has been developed to approximate the effect of model uncertainty, sensor noise, data rate, and thrust profile error on the expected estimate of collected sample mass.

Carson, John↗

Low depth quantum circuits for Ising models

A scheme for measuring complex temperature partition functions of Ising models is introduced. Two applications of this scheme are presented. First, through appropriate Wick rotations, those amplitudes can be analytically continued to yield estimates for partition functions of Ising models. Bounds on the estimated error are provided through a central-limit theorem whose validity extends beyond the present context; it holds for example for estimations of the Jones polynomial. The kind of state preparations and measurements involved in this application can be made independent of the system size or the parameters of the system being simulated. Second, the scheme allows to accurately estimate non-trivial invariants of links. Another result concerns the computational power of estimations of partition functions for real temperature classical ferromagnetic Ising models. We provide conditions under which estimating such partition functions allows to reconstruct scattering amplitudes of quantum circuits, making the problem BQP-hard. We also show fidelity overlaps for ground states of quantum Hamiltonians, which serve as a witness to quantum phase transitions, can be estimated from classical Ising model partition functions. Finally, we discuss how accurate corner magnetisation measurements on thermal states of two-dimensional Ising models lead to fully polynomial random approximation schemes (FPRAS) for the partition function.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Logical error rates for the surface code under a mixed coherent and stochastic circuit-level noise model inspired by trapped ions

With fault-tolerant quantum computing (FTQC) on the horizon, it is critical to understand sources of logical errors in plausible hardware implementations of quantum error-correcting codes. Detailed error modeling of computational instructions on particular FTQC architectures will enable the better prediction of error propagation in FT-encoded quantum circuits while revealing where greater attention is needed in hardware design. In this work, we consider logical error rates for the surface code implemented on a hypothetical grid-based trapped-ion quantum charge-coupled device architecture. Specifically, we construct logical channels for the idling surface code and examine its diamond error under a mixed coherent and stochastic circuit-level noise model inspired by trapped ions. We include the coherent dephasing noise that is known to accumulate during physical qubit idling and transport in these systems, determining idling and transport durations using the time-resolved output of an open-source trapped-ion surface code compiler. To estimate expectation values of logical Pauli observables following hardware circuits containing non-Clifford sources of noise, we utilize a Monte Carlo technique to sample from an underlying quasiprobability distribution of Clifford circuits that we independently simulate in a phase-sensitive fashion. We verify error suppression up to code distance 𝑑 = 11 at coherent dephasing rates near and below those of current-generation trapped-ion quantum computers and find that logical error rates align with those of analogous fully stochastic simulations in this regime. Exploring higher dephasing rates at 𝑑 = 3−5, we find evidence for growing coherent rotations about all three logical Pauli axes, increased diagonal logical error process matrix elements relative to those of stochastic simulations, and a reduced dephasing rate threshold. Overall, our work paves a way toward realistic hardware emulation of small fault-tolerant quantum processes, e.g., members of an FTQC instruction set.

Quantum benchmarking↗

Graph Neural Networks for Parameterized Quantum Circuits Expressibility Estimation (Rev.1)

Parameterized quantum circuits (PQCs) are fundamental to quantum machine learning (QML), quantum optimization, and variational quantum algorithms (VQAs). The expressibility of PQCs is a measure that determines their capability to harness the full potential of the quantum state space. It is thus a crucial guidepost to know when selecting a particular PQC ansatz. However, the existing technique for expressibility computation through statistical estimation requires a large number of samples, which poses significant challenges due to time and computational resource constraints. This paper introduces a novel approach for expressibility estimation of PQCs using Graph Neural Networks (GNNs). We demonstrate the predictive power of our GNN model with a dataset consisting of 25,000 samples from the noiseless IBM QASM Simulator and 12,000 samples from three distinct noisy quantum backends. The model accurately estimates expressibility, with root mean square errors (RMSE) of 0.05 and 0.06 for the noiseless and noisy backends, respectively. We compare our model’s predictions with reference circuits from Sim et al. and IBM Qiskit’s hardwareefficient ansatz sets to further evaluate our model’s performance. Our experimental evaluation in noiseless and noisy scenarios reveals a close alignment with ground truth expressibility values, highlighting the model’s efficacy. Moreover, our model exhibits promising extrapolation capabilities, predicting expressibility values with low RMSE for out-of-range qubit circuits trained solely on only up to 5-qubit circuit sets. This work thus provides a reliable means of efficiently evaluating the expressibility of diverse PQCs on noiseless simulators and hardware.

97 MATHEMATICS AND COMPUTING↗

Empirically-calibrated H100 node power models for accurate AI training energy estimation

Accurately quantifying the energy use of artificial intelligence (AI) training is critical for infrastructure planning, carbon accounting, and sustainable data center operation, but few studies have directly measured the power consumption of production workloads on contemporary hardware. By combining empirical measurements from Brookhaven National Laboratory during AI training on 8-graphics-processing-unit H100 systems with open-source benchmarking data, we develop statistical models relating computational intensity to node-level power consumption. We measure the gap between manufacturer-rated thermal design power (TDP) and actual power demand during AI training. Our analysis reveals that even computationally intensive workloads operate at only 76% of the 10.2 kW TDP rating. Our architecture-specific model, calibrated to floating-point operations, predicts energy consumption with 11.4% mean absolute percentage error, significantly outperforming TDP-based approaches (27%–37% error). We identified distinct power signatures between transformer and convolutional neural network architectures, with transformers showing characteristic fluctuations that may impact grid stability. These results provide a measurement-grounded basis for improving AI training energy estimates, enabling more reliable infrastructure sizing, cost projections, and environmental impact assessments.

Newkirk, Alex C↗

Advanced Small Modular Reactor Economics Model Development

The US Department of Energy Office of Nuclear Energy’s Advanced Small Modular Reactor (SMR) research and development activities focus on four key areas: Developing assessment methods for evaluating advanced SMR technologies and characteristics; and Developing and testing of materials, fuels and fabrication techniques; and Resolving key regulatory issues identified by US Nuclear Regulatory Commission and industry; and Developing advanced instrumentation and controls and human-machine interfaces. This report focuses on development of assessment methods to evaluate advanced SMR technologies and characteristics. Specifically, this report describes the expansion and application of the economic modeling effort at Oak Ridge National Laboratory. Analysis of the current modeling methods shows that one of the primary concerns for the modeling effort is the handling of uncertainty in cost estimates. Monte Carlo–based methods are commonly used to handle uncertainty, especially when implemented by a stand-alone script within a program such as Python or MATLAB. However, a script-based model requires each potential user to have access to a compiler and an executable capable of handling the script. Making the model accessible to multiple independent analysts is best accomplished by implementing the model in a common computing tool such as Microsoft Excel. Excel is readily available and accessible to most system analysts, but it is not designed for straightforward implementation of a Monte Carlo–based method. Using a Monte Carlo algorithm requires in-spreadsheet scripting and statistical analyses or the use of add-ons such as Crystal Ball. An alternative method uses propagation of error calculations in the existing Excel-based system to estimate system cost uncertainty. This method has the advantage of using Microsoft Excel as is, but it requires the use of simplifying assumptions. These assumptions do not necessarily bring into question the analytical results. In fact, the analysis shows that the propagation of error method introduces essentially negligible error, especially when compared to the uncertainty associated with some of the estimates themselves. The results of these uncertainty analyses generally quantify and identify the sources of uncertainty in the overall cost estimation. The obvious generalization—that capital cost uncertainty is the main driver—can be shown to be an accurate generalization for the current state of reactor cost analysis. However, the detailed analysis on a component-by-component basis helps to demonstrate which components would benefit most from research and development to decrease the uncertainty, as well as which components would benefit from research and development to decrease the absolute cost.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Quantifying the uncertainty introduced by discretization and time-averaging in two-fluid model predictions

The two-fluid model (TFM) has become a tool for the design and troubleshooting of industrial fluidized bed reactors. To use TFM for scale up with confidence, the uncertainty in its predictions must be quantified. Here, we study two sources of uncertainty: discretization and time-averaging. First, we show that successive grid refinement may not yield grid-independent transient quantities, including cross-section–averaged quantities. Successive grid refinement would yield grid-independent time-averaged quantities on sufficiently fine grids. A Richardson extrapolation can then be used to estimate the discretization error, and the grid convergence index gives an estimate of the uncertainty. Richardson extrapolation may not work for industrial-scale simulations that use coarse grids. We present an alternative method for coarse grids and assess its ability to estimate the discretization error. Second, we assess two methods (autocorrelation and binning) and find that the autocorrelation method is more reliable for estimating the uncertainty introduced by time-averaging TFM data.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Local prediction of Laser Powder Bed Fusion porosity by short-wave infrared imaging thermal feature porosity probability maps

We report that local thermal history can significantly vary in parts during metal Additive Manufacturing (AM), leading to local defects. However, the sequential layer-by-layer nature of AM facilitates in-situ part voxelmetric observations that can be used to detect and correct these defects for part qualification and quality control. The challenge is to relate this local radiometric data with local defect information to estimate process error likelihood in future builds. This paper uses a Short-Wave Infrared (SWIR) camera to record the temperature history for parts manufactured with Laser Powder Bed Fusion (LPBF) processes. The porosity from a cylindrical specimen is measured by ex-situ micro-computed tomography (μCT). Specimen data from the SWIR camera, combined with the μCT data, are used to generate thermal feature-based porosity probability maps. The porosity predictions made by various SWIR thermal feature-porosity probability maps of a specimen with a complex geometry are scored against the true porosity obtained via μCT. The receiver operating characteristic curves constructed from the predictions for the complex sample demonstrate the porosity probability mapping methodology’s potential for in-situ based porosity detection.

36 MATERIALS SCIENCE↗