SEARCH · Search NASA

Results for “prediction 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.

SourceResearch connectionFrom yearThrough year

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

At least 235 records · Page 13

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Development of advanced machine learning models for analysis of plutonium surrogate optical emission spectra

This work investigates and applies machine learning paradigms seldom seen in analytical spectroscopy for quantification of gallium in cerium matrices via processing of laser-plasma spectra. Ensemble regressions, support vector machine regressions, Gaussian kernel regressions, and artificial neural network techniques are trained and tested on cerium-gallium pellet spectra. A thorough hyperparameter optimization experiment is conducted initially to determine the best design features for each model. The optimized models are evaluated for sensitivity and precision using the limit of detection (LoD) and root mean-squared error of prediction (RMSEP) metrics, respectively. Gaussian kernel regression yields the superlative predictive model with an RMSEP of 0.33% and an LoD of 0.015% for quantification of Ga in a Ce matrix. This study concludes that these machine learning methods could yield robust prediction models for rapid quality control analysis of plutonium alloys.

Rao, Ashwin P. (ORCID:0000000319312568)↗

DOE OSTI2023

Establishing Isotopic Measurement Capabilities using Laser-Induced Breakdown Spectroscopy for the Molten Salt Reactor Campaign

Proof-of-principle H isotope ratios of aqueous aerosol systems and liquid droplets on filter paper have been measured using laser-induced breakdown spectroscopy (LIBS) with root mean square error of prediction values down to 1.9%. Molten salt reactors (MSRs) will consist of a complex chemical and radiological system consistently producing new fission products as the reactor operates. Some of these fission products and/or their daughter species will leave the salt in the reactor off-gas. Monitoring the composition of the off-gas, as well as the salt itself, is important for monitoring reactor performance, including burnup, corrosion, and the concentration of impurities. Tritium is of concern for MSRs because it will be produced from the irradiation of key salt constituents, including Li and Be. LIBS offers an avenue for in situ salt and off-gas monitoring by firing a laser onto or into the sample stream to generate a plasma. The plasma light can be monitored to measure an elemental fingerprint of the sample. Extending this analysis to include isotopic ratios offers a critical expansion of the in situ monitoring capabilities being developed. This study demonstrates the expansion of Oak Ridge National Laboratory’s LIBS capabilities to monitor isotopes and shows how simple calibrations may provide rapid semiquantitative models.

07 ISOTOPE AND RADIATION SOURCES↗

DOE OSTI2023

Smart Pixels: towards on-sensor inference of charged particle track parameters and uncertainties

The combinatorics of track seeding has long been a computational bottleneck for triggering and offline computing in High Energy Physics (HEP), and remains so for the HL-LHC. Next-generation pixel sensors will be sufficiently fine-grained to determine angular information of the charged particle passing through from pixel-cluster properties. This detector technology immediately improves the situation for offline tracking, but any major improvements in physics reach are unrealized since they are dominated by lowest-level hardware trigger acceptance. We will demonstrate track angle and hit position prediction, including errors, using a mixture density network within a single layer of silicon as well as the progress towards and status of implementing the neural network in hardware on both FPGAs and ASICs.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2026

Empirical Comparison of Machine Learning Approaches for Black-Box Modeling of Power Conversion System Dynamics

Inverter-based resources are key components in modern power systems, but accurately modeling their complex behavior can be challenging. Standard, generic converter models often oversimplify inverter dynamics, leading to significant errors in predicting performance. In this work, we compare several data-driven machine learning (ML) approaches for inverter modeling, performing experiments on power conversion systems, systematically varying input conditions, and recording the resulting voltages and currents. The ML models were then trained on this measured data to capture the inverter's dynamic response and to predict the inverter's output current. A performance comparison between the four ML models under study is conducted, laying the foundation for future work on hardware implementation for real-time inference.

30 DIRECT ENERGY CONVERSION↗

DOE OSTI2020

Reconciling Chord Length Distributions and Area Distributions for Fields of Fractal Cumulus Clouds

While the total cover of broken cloud fields can in principle be obtained from one-dimensional measurements, the cloud size distribution normally differs between two-dimensional (area) and one-dimensional retrieval (chord length) methods. In this study, we use output from high-resolution Large Eddy Simulations to generate a transfer function between the two. We retrieve chord lengths and areas for many clouds, and plot the one as a function of the other, and vice versa. We find that the cloud area distribution conditional on the chord length behaves like a gamma distribution with well-behaved parameters, with a mean μ=1.1L and a shape parameter β=L−0.645. Using this information, we are able to generate a transfer function that can adjust the chord length distribution so that it comes much closer to the cloud area distribution. Our transfer function improves the error in predicting the mean cloud size, and is performs without strong biases for smaller sample sizes. However, we find that the method is still has difficulties in accurately predicting the frequency of occurrence of the largest cloud sizes.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2020

Identification of Promising Alternative Mono-Alcohol Fuel Blend Components for Spark Ignition Engines

Alcohols are attractive fuel blendstocks for spark ignition engines due to their high octane values and potentially positive influence on performance and emission. Although methanol, ethanol, and butanol have been widely studied, other biomass-derived alcohols may have similar or better properties. However, it is not feasible to experimentally investigate the fuel potential of every molecule. The goals of this study were to develop a methodology for rapid screening of a fuel property database for mono-alcohols and to identify alcohols with the potential of blending to produce advantaged motor gasolines. A database was developed with 13 fuel properties of all saturated C1–C10 mono-alcohols. A decision framework was used to evaluate alcohols suitable for blending in gasoline for spark ignition engines in two scenarios: low-range (up to 15 vol%) blends and high-range (greater than 40 vol%) blends. The low-range blend cases resulted in the identification of 48 alcohols. In the case of high-range blending, only six alcohols were found to be suitable. This is the first study to systematically evaluate all C1–C10 saturated alcohols for blending with gasoline using relevant fuel properties. A novel aspect of this study is the evaluation of the influence of errors in predicted property values. These scenario screenings focus attention on a smaller number of promising candidate molecules, and the approach could be modified for other classes of fuel molecules, engine types, and fuel blending goals.

42 ENGINEERING↗

DOE OSTI2023

Partial Least Squares, Experimental Design, and Near-Infrared Spectrophotometry for the Remote Quantification of Nitric Acid Concentration and Temperature

Near-infrared spectrophotometry and partial least squares regression (PLSR) were evaluated to create a pleasantly simple yet effective approach for measuring HNO3 concentration with varying temperature levels. A training set, which covered HNO3 concentrations (0.1–8 M) and temperature (10–40 °C), was selected using a D-optimal design to minimize the number of samples required in the calibration set for PLSR analysis. The top D-optimal-selected PLSR models had root mean squared error of prediction values of 1.4% for HNO 3 and 4.0% for temperature. The PLSR models built from spectra collected on static samples were validated against flow tests including HNO 3 concentration and temperature gradients to test abnormal conditions (e.g., bubbles) and the model performance between sample points in the factor space. Based on cross-validation and prediction modeling statistics, the designed near-infrared absorption approach can provide remote, quantitative analysis of HNO 3 concentration and temperature for production-oriented applications in facilities where laser safety challenges would inhibit the implementation of other optical techniques (e.g., Raman spectroscopy) and in which space, time, and/or resources are constrained. The experimental design approach effectively minimized the number of samples in the training set and maintained or improved PLSR model performance, which makes the described chemometric approach more amenable to nuclear field applications.

07 ISOTOPE AND RADIATION SOURCES↗

DOE OSTI2024

Vehicle Position Detection Based on Machine Learning Algorithms in Dynamic Wireless Charging

Dynamic wireless charging (DWC) has emerged as a viable approach to mitigate range anxiety by ensuring continuous and uninterrupted charging for electric vehicles in motion. DWC systems rely on the length of the transmitter, which can be categorized into long-track transmitters and segmented coil arrays. The segmented coil array, favored for its heightened efficiency and reduced electromagnetic interference, stands out as the preferred option. However, in such DWC systems, the need arises to detect the vehicle’s position, specifically to activate the transmitter coils aligned with the receiver pad and de-energize uncoupled transmitter coils. This paper introduces various machine learning algorithms for precise vehicle position determination, accommodating diverse ground clearances of electric vehicles and various speeds. Through testing eight different machine learning algorithms and comparing the results, the random forest algorithm emerged as superior, displaying the lowest error in predicting the actual position.

47 OTHER INSTRUMENTATION↗

DOE OSTI2022

Advanced Finite-Volume Numerics and Source Term Assumptions for Kernel and G-Equation Modelling of Propane/Air Flames

Here G-Equation models represent propagating flame fronts with an implicit two-dimensional surface representation (level-set). Level-set methods are fast, as transport source terms for the implicit surface can be solved with finite-volume operators on the finite-volume domain, without having to build the actual surface. However, they include approximations whose practical effects are not properly understood. In this study, we improved the numerics of the FRESCO CFD code’s G-Equation solver and developed a new method to simulate kernel growth using signed distance functions and the analytical sphere-mesh overlap. We analyzed their role for simulating propane/air flames, using three well-established constant-volume configurations: a one-dimensional, freely propagating laminar flame; a disc-shaped, constant-volume swirl combustor; and torch-jet flame development through an orifice from a two-chamber device. We tested the explicit (sub-cycled) vs. implicit formulation for the standard transport operators (advection, diffusion, compressibility). In addition to the accurate flame swept-volume method for chemistry and species source term, we developed a more accurate estimator for the burnt/unburnt split cell composition. Then, we developed a signed-distance-function (SDF) based method which provides a more stable reinitialization of the level-set field at every time-step. We found that simplifying assumptions common to several G-Equation implementations, for straightforward terms such as compressibility and advection, lead to large errors in predicting the propagation of even laminar flames, with deviations up to ~300% in simulated vs. formulated flame speed. Conversely, the enhanced numerics enabled through the SDF field reinitialization and improved chemistry source term improve simulation stability and smooth flame propagation even with significantly larger solver time-steps.

42 ENGINEERING↗

DOE OSTI2023

Development of a Supercharged Octane Number and a Supercharged Octane Index

Gasoline knock resistance is characterized by the Research and Motor Octane Number (RON and MON), which are rated on the CFR octane rating engine at naturally aspirated conditions. However, modern automotive downsized boosted spark ignition (SI) engines generally operate at higher cylinder pressures and lower temperatures relative to the RON and MON tests. Using the naturally aspirated RON and MON ratings, the octane index (OI) characterizes the knock resistance of gasolines under boosted operation by linearly extrapolating into boosted “beyond RON” conditions via RON, MON, and a linear regression K factor. Using OI solely based on naturally aspirated RON and MON tests to extrapolate into boosted conditions can lead to significant errors in predicting boosted knock resistance between gasolines due to non-linear changes in autoignition and knocking characteristics with increasing pressure conditions. Here, a new “Supercharged Octane Number” (SON) method was developed on the CFR engine at increased intake pressures, which improved the correlation to boosted knock-limited automotive SI engine data over RON for several surrogate fuels and gasolines, including five “Co-Optima” RON 98 fuels and an E10 regular grade gasoline. Furthermore, the conventional OI was extended to a newly introduced Supercharged Octane Index (OI S ) based on SON and RON, which significantly improved the correlation to fuel knock resistance measurements from modern boosted SI engine knock-limited spark advance tests. This demonstrated the first proof of concept of a SON and OI S to better characterize a fuel’s knock resistance in modern boosted SI engines.

42 ENGINEERING↗

DOE OSTI2020

Extraction of mechanical properties of materials through deep learning from instrumented indentation

Instrumented indentation has been developed and widely utilized as one of the most versatile and practical means of extracting mechanical properties of materials. This method is particularly desirable for those applications where it is difficult to experimentally determine the mechanical properties using stress–strain data obtained from coupon specimens. Such applications include material processing and manufacturing of small and large engineering components and structures involving the following: three dimensional (3D) printing, thin-film and multilayered structures, and integrated manufacturing of materials for coupled mechanical and functional properties. Here, we utilize the latest developments in neural networks, including a multifidelity approach whereby deep-learning algorithms are trained to extract elastoplastic properties of metals an d alloys from instrumented indentation results using multiple datasets for desired levels of improved accuracy. We have established algorithms for solving inverse problems by recourse to single, dual, and multiple indentation and demonstrate that these algorithms significantly outperform traditional brute force computations and function-fitting methods. Moreover, we present several multifidelity approaches specifically for solving the inverse indentation problem which 1) significantly reduce the number of high-fidelity datasets required to achieve a given level of accuracy, 2) utilize known physical and scaling laws to improve training efficiency and accuracy, and 3) integrate simulation and experimental data for training disparate datasets to learn and minimize systematic errors. The predictive capabilities and advantages of these multifidelity methods have been assessed by direct comparisons with experimental results for indentation for different commercial alloys, including two wrought aluminum alloys and several 3D printed titanium alloys.

3D printed materials, stress-strain behavior, mult↗

DOE OSTI2023

Simulation-Based Recovery Action Analysis Using the EMRALD Dynamic Risk Assessment Tool

Recovery human action is defined as the action that prevents deviant conditions from producing unwanted effects. Analyzing recovery actions has been a critical part in human reliability analysis (HRA). However, there are a couple of limitations to treating recovery actions using only the current HRA methods available. Representatively, the existing recovery analysis does not specifically consider recovery actions as they have occurred in actual nuclear power plants (NPPs). To handle the challenges in the existing recovery analyses, this study suggests a way to analyze recovery actions under a dynamic HRA method, the Procedure-based Risk Investigation MEthod-Human Reliability Analysis (PRIME-HRA) method. The PRIME-HRA method suggests a way on how to develop dynamic simulation models using dynamic risk assessment tools such as the Event Modeling Risk Assessment Using Linked Diagram (EMRALD) [1] and the Human Unimodel for Nuclear Technology to Enhance Reliability (HUNTER) [2]. EMRALD and HUNTER are the dynamic probabilistic risk assessment and HRA tools developed at Idaho National Laboratory. In this paper, differences on analyzing recovery actions in the Technique for Human Error-Rate Prediction (THERP), the Cause-Based Decision Tree (CBDT) and the Korean Standard HRA (K-HRA) and challenges of these approaches are introduced. How we have developed the PRIME-HRA is also introduced in this paper. Then, the proposed approach to analyzing recovery human actions in dynamic context is partially discussed with an example.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2023

Simulation-Based Recovery Action Analysis Using the EMRALD Dynamic Risk Assessment Tool

A recovery action is defined as the action that prevents deviant conditions from producing unwanted effects. It generally indicates a kind of countermeasure performed in response to a failure of human action. The recovery actions especially play an important role in complex systems like nuclear power plants (NPPs), which consist of highly sophisticated controllers to ensure that desired performance and safety must be achieved and maintained. This is because a combination of human error and its recovery failure may be able to cause a catastrophic effect on a system. Analyzing recovery actions has been a critical part of HRA, which is a technique to evaluate human errors and provide human error probabilities (HEPs) for application in probabilistic safety assessment (PSA). If recovery actions are not adequately analyzed and applied to PSA models, the PSA results may be under-estimated or be not able to reasonably account for the failure of human actions in the context of PSA. For this reason, some regulatory documents such as ASME/ANS RA-Sb-2013 by the American Society for Mechanical Engineers and the American Nuclear Society and NUREG-1792 by U.S. Nuclear Regulatory Commission have emphasized the importance of recovery analysis within the HRA. A couple of existing HRA methods, such as the Technique for Human Error-Rate Prediction (THERP), the Cause-Based Decision Tree (CBDT), and the Korean Standard HRA (K-HRA), have respectively suggested their own approaches to the HRA recovery analysis. However, there are a couple of limitations to treating recovery actions using only the current HRA methods available. The biggest limitation is that the existing recovery analysis does not explicitly consider a variety of recovery action types and recovery sequences as they occur in actual NPPs. To handle the limitations of existing recovery analysis, this study proposes a simulation-based recovery analysis method using the Event Modeling Risk Assessment Using Linked Diagram (EMRALD) software. The EMRALD software is a dynamic simulation tool for PSA. It supports realistic and dynamic modeling of human actions as they would be performed at NPPs. It is also favorable to simultaneously model the specific moment at which an action is performed, the time it takes to perform the action, and the failure probability of that action. In this paper, a detailed methodology for modeling recovery actions in the simulation platform is proposed with a couple of examples. Then, outputs from the simulation are discussed as reviewing if this novel approach can complement the challenges of existing recovery analyses.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2023

Smart Pixels: towards on-sensor inference of charged particle track parameters and uncertainties

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2023

Smartpixels: Towards on-sensor inference of charged particle track parameters and uncertainties

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2020

ScaleML: Machine Learning based Heap Memory Object Scaling Prediction

Memory subsystem contributes 28-40% of total energy consumption. Several studies investigated energy prediction and consumption via profiling memory object access patterns. However, such profiling leads to higher energy consumption due to intense memory object-level profiling to achieve high prediction accuracy. Further, memory object access pattern prediction has been considered through analyzing the variation between memory object access patterns, referred to as scaling rate. The existing techniques for scaling rate prediction, such as Linear Scaling Rate (LSR), suffer from a high error rate in prediction with changes in access patterns, which leads to a high error rate of energy consumption prediction. In this paper, we compare and evaluate several memory object access pattern prediction models including LSR and machine learning (ML) models. Further, we propose SCALEML, a heap memory object scaling rate prediction mechanism that employs an ML model to achieve high access pattern prediction accuracy with variations in memory object access patterns. We evaluate SCALEML using various application benchmarks. The experimental results show that SCALEML achieves about 20% higher accuracy than the LSR model for predicting the scaling rate of object access patterns and energy estimation.

Park, Joongeon↗

DOE OSTI2021

A fast multi-fidelity method with uncertainty quantification for complex data correlations: Application to vortex-induced vibrations of marine risers

Here we develop a fast multi-fidelity modeling method for very complex correlations between high- and low-fidelity data by working in modal space to extract the proper correlation function. We apply this method to infer the amplitude of motion of a flexible marine riser in cross-flow, subject to vortex-induced vibrations (VIV). VIV are driven by an absolute instability in the flow, which imposes a frequency (Strouhal) law that requires a matching with the impedance of the structure; this matching is easily achieved because of the rapid parametric variation of the added mass force. As a result, the wavenumber of the riser spatial response is within narrow bands of uncertainty. Hence, an error in wavenumber prediction can cause significant phase-related errors in the shape of the amplitude of response along the riser, rendering correlation between low- and high-fidelity data very complex. Working in modal space as outlined herein, dense data from low-fidelity data, provided by the semi-empirical computer code VIVA, can correlate in modal space with few high-fidelity data, obtained from experiments or fully-resolved CFD simulations, to correct both phase and amplitude and provide predictions that agree very well overall with the correct shape of the amplitude response. We also quantify the uncertainty in the prediction using Bayesian modeling and exploit this uncertainty to formulate an active learning strategy for the best possible location of the sensors providing the high fidelity measurements.

42 ENGINEERING↗

DOE OSTI2020

The Power Curve Working Group's assessment of wind turbine power performance prediction methods

Wind turbine power production deviates from the reference power curve in real-world atmospheric conditions. Correctly predicting turbine power performance requires models to be validated for a wide range of wind turbines using inflow in different locations. The Share-3 exercise is the most recent intelligence-sharing exercise of the Power Curve Working Group, which aims to advance the modeling of turbine performance. The goal of the exercise is to search for modeling methods that reduce error and uncertainty in power prediction when wind shear and turbulence digress from design conditions. Herein, we analyze data from 55 wind turbine power performance tests from nine contributing organizations with statistical tests to quantify the skills of the prediction-correction methods. We assess the accuracy and precision of four proposed trial methods against the baseline method, which uses the conventional definition of a power curve with wind speed and air density at hub height. The trial methods reduce power-production prediction errors compared to the baseline method at high wind speeds, which contribute heavily to power production; however, the trial methods fail to significantly reduce prediction uncertainty in most meteorological conditions. For the meteorological conditions when a wind turbine produces less than the power its reference power curve suggests, using power deviation matrices leads to more accurate power prediction. We also determine that for more than half of the submissions, the data set has a large influence on the effectiveness of a trial method. Overall, this work affirms the value of data-sharing efforts in advancing power curve modeling and establishes the groundwork for future collaborations.

17 WIND ENERGY↗