Search NASA⌕ Search

SEARCH · Search NASA

Results for “model accuracy”

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
DOE OSTI2021

Selecting durable building envelope systems with machine learning assisted hygrothermal simulations database

Hygrothermal simulations provide insight into the energy performance and moisture durability of building envelope components under dynamic conditions. The inputs required for hygrothermal simulations are extensive, and carrying out simulations and analyses requires expert knowledge. An expert system, the Building Science Advisor (BSA), has been developed to predict the performance and select the energy-efficient and durable building envelope systems for different climates. The BSA consists of decision rules based on expert opinions and thousands of parametric simulation results for selected wall systems. The number of potential wall systems results in millions, too many to simulate all of them. We present how machine learning can help predict durability data, such as mold growth, while minimizing the number of simulations needed to run. The simulation results are used for training and validation of machine learning tools for predicting wall durability. We tested Artificial Neural Network (ANN) and Gradient Boosted Decision Trees (GBDT) for their applicability and model accuracy. Models developed with both methods showed adequate prediction performance (root mean square error of 0.195 and 0.209, respectively). Finally, we introduce how the information supports guidance for envelope design via an easy-to-use web-based tool that does not require the end-user to run hygrothermal simulations.

Salonvaara, Mikael↗
NASA NTRS2023

Comparative Accuracies of Models for Drag Prediction During Geomagnetically Disturbed Periods: A First Principles Model Versus Empirical Models

We examine the accuracy of density prediction by the first principles model Thermosphere Ionosphere Electrodynamics General Circulation Model (TIEGCM) developed by the National Center for Atmospheric Research and compare it to the accuracy of three empirical models: Jacchia 71, the Naval Research Laboratory Mass Spectrometer Incoherent Scatter Extended 2000 (NRLMSIS), Jacchia 1971, and Jacchia-Bowman 2008. Comparisons are made for three large storms: the October 2003 storm, the March 2013 storm, and the March 2015 storm. To evaluate the accuracy of these models we use tracking data for nine space objects in low Earth orbit. Additionally, we evaluate the accuracy of the TIEGCM and NRLMSIS with data from high precision accelerometers on the Challenging Minisatellite Payload (CHAMP) and Gravity field and Circulation Explorer (GOCE) satellites. The goal is to assess the use of a first principles model as a potential tool for forecasting satellite drag during large magnetic storms. For the storms considered, we found the TIEGCM, JB2008, and NRLMSIS models to be substantially more accurate than the Jacchia 71 model. The accuracies of the TIEGCM and JB2008 models were similar, but overall, the TIEGCM was more accurate. We found smaller differences for TIEGCM versus CHAMP than for NRLMIS for the Halloween Storm, and smaller differences than results published for JB2008 and the assimilative model HASDM. The empirical models are at present more practical for operational purposes, but the TIEGCM, developed as a research model, with a greater focus on operational use offers the potential for improved utility during stressing conditions.

R. L. Walterscheid↗
NASA NTRS1976

Method of discrete modeling and its application to estimation of TF30 engine variables

A method of discrete modeling is presented that effectively isolates steady state model accuracy from dynamic model accuracy. The steady state model may be generated from the engine design equations with any desired degree of accuracy. The dynamic model is generated by applying a step disturbance of a manipulated variable to an open loop engine simulation. The sampled response of the variable is combined with the steady state model's response to form a set of weighting factors. These weighting factors are then used to weight past values of the manipulated variable, thus forming the dynamic model. The method is used to estimate various TF30-P-3 engine variables. A dynamic trim function is developed to compensate for the dynamic nonlinearities of the variables as well as for inaccuracies in dynamic definition. The trim function is shown to be realted to the square root of the sum of the squares of the weighting factors obtained at various engine operating conditions. Finally, the estimation of variables without dynamic modeling is discussed.

Arpasi, D. J.↗
NASA NTRS1978

The accuracy of geopotential models

The accuracy of two recent geopotential models, GEM 7 and GEM 8, is evaluated and is found to be about 4.3 m with respect to the global geoid surface for GEM 7 and 3.9 m for GEM 8. The accuracies are root mean square values obtained by the use of 400 coefficients for GEM 7 and 706 coefficients for GEM 8. Independent observations used in the evaluation include 159 lumped coefficients from 35 resonant orbits, two sets of fields derived from optical-only and laser-only data, sets of zonal and resonant coefficients, and geoid undulations. The ratio of estimated commission to formal error in GEM 7 and GEM 8 ranges from 2 to 5. Several other recent geopotential models are examined.

Wagner, C. A.↗
NASA NTRS1983

Subpixel registration accuracy and modelling

An outline of methods for subpixel registration accuracy and modelling is presented. Consideration of the following questions is emphasized: how accurately can a LANDSAT image be registered to a reference image, how can subpixel accuracy be achieved, what factors affect registration accuracy, how should reference images be formed, and how can various algorithms be evaluated.

Kanal, L. N.↗
DOE OSTI2021

Interpreting Write Performance of Supercomputer I/O Systems with Regression Models

This work seeks to advance the state of the art in HPC I/O performance analysis and interpretation. In particular, we demonstrate effective techniques to: (1) model output performance in the presence of I/O interference from production loads; (2) build features from write patterns and key parameters of the system architecture and configurations; (3) employ suitable machine learning algorithms to improve model accuracy. We train models with five popular regression algorithms and conduct experiments on two distinct production HPC platforms. We find that the lasso and random forest models predict output performance with high accuracy on both of the target systems. We also explore use of the models to guide adaptation in I/O middleware systems, and show potential for improvements of at least 15% from model-guided adaptation on 70% of samples, and improvements up to 10× on some samples for both of the target systems.

Xie, Bing↗
DOE OSTI2021

Active learning strategy for high fidelity short-term data-driven building energy forecasting

The quality of a data-driven model is heavily dependent on the quality of data. Data from building operation often have data bias problems, which means that the data sample is collected in a way that some members of the intended data population are less likely to be included than others. Data-driven energy forecasting models built on such data hence are biased and could lead to large forecasting errors. Active learning—an effective method to defying data bias—is rarely studied or applied in the area of data-driven building energy forecasting modeling. This paper attempts to fill this gap and explores the application of active learning in data-driven building energy forecasting. The developed strategy in this paper efficiently generate informative training data within a time budget and uses block design to passively consider weather disturbances. The developed active learning strategy is applied and evaluated in both virtual and real-building testbeds against traditional data-driven methods. Via these virtual and real-building evaluation cases, we have demonstrated that the data bias problem typically exists in building operation data is resolved by applying the developed active learning strategy. Furthermore, building energy forecasting models trained from data generated from the active learning strategy have shown improved performances in both model accuracy and model extendibility perspectives. The effectiveness of the block design module is also validated to effectively consider the impact of weather conditions on active learning design.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS1992

Thermal modelling of an AMTEC recirculating cell

A modeling program was developed to determine the impact of various design parameters on the operation of an AMTEC system. Temperature profiles generated by the modeling program were compared to actual experimental data to verify the model accuracy. The model was then extended to predict the impact of device design on operational performance. The effect of heat loss from the liquid sodium supply end was studied for this paper.

Suitor, J. W.↗
NASA NTRS1995

Flight assessment of the onboard propulsion system model for the Performance Seeking Control algorithm on an F-15 aircraft

Performance Seeking Control (PSC), an onboard, adaptive, real-time optimization algorithm, relies upon an onboard propulsion system model. Flight results illustrated propulsion system performance improvements as calculated by the model. These improvements were subject to uncertainty arising from modeling error. Thus to quantify uncertainty in the PSC performance improvements, modeling accuracy must be assessed. A flight test approach to verify PSC-predicted increases in thrust (FNP) and absolute levels of fan stall margin is developed and applied to flight test data. Application of the excess thrust technique shows that increases of FNP agree to within 3 percent of full-scale measurements for most conditions. Accuracy to these levels is significant because uncertainty bands may now be applied to the performance improvements provided by PSC. Assessment of PSC fan stall margin modeling accuracy was completed with analysis of in-flight stall tests. Results indicate that the model overestimates the stall margin by between 5 to 10 percent. Because PSC achieves performance gains by using available stall margin, this overestimation may represent performance improvements to be recovered with increased modeling accuracy. Assessment of thrust and stall margin modeling accuracy provides a critical piece for a comprehensive understanding of PSC's capabilities and limitations.

Orme, John S.↗
DOE OSTI2021

High-resolution hourly surrogate modeling framework for physics-based large-scale building stock modeling

Surrogate modeling can play a key role in reducing high computational burdens for large-scale physics-based modeling and uncertainty quantification. With the rapid development of large-scale building stock energy modeling, surrogate modeling has also begun to be widely applied in this field; however, most existing surrogate models lack hourly time resolution for regional-scale modeling, which is essential for understanding building demand profiles and grid impacts. Further, there is generally a lack of necessary data and feature engineering frameworks specific to building modeling for efficiently managing large datasets and complex computations. Furthermore, this paper proposes a modeling framework for large-scale (city-/region-scale), high-resolution, high-fidelity surrogate building stock energy models. Our developed framework consists of six modules: (1) building stock energy modeling (ComStock TM and ResStock TM ), (2) data engineering for large simulation data, (3) high performance computing workflow, (4) feature engineering, (5) machine learning model development, and (6) model performance evaluation. Two case studies apply the developed framework in both residential and commercial building stock analysis to demonstrate its computational efficiency and surrogate modeling accuracies. Results show that surrogate models, when efficiently trained using the HPC workflow module, reach a high level of modeling accuracy for two case studies.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

Improving scalability of parallel CNN training by adaptively adjusting parameter update frequency

Synchronous SGD with data parallelism, the most popular parallelization strategy for CNN training, suffers from the expensive communication cost of averaging gradients among all workers. The iterative parameter updates of SGD cause frequent communications and it becomes the performance bottleneck. In this paper, we propose a lazy parameter update algorithm that adaptively adjusts the parameter update frequency to address the expensive communication cost issue. Our algorithm accumulates the gradients if the difference of the accumulated gradients and the latest gradients is sufficiently small. Here, the less frequent parameter updates reduce the per-iteration communication cost while maintaining the model accuracy. Our experimental results demonstrate that the lazy update method remarkably improves the scalability while maintaining the model accuracy. For ResNet50 training on ImageNet, the proposed algorithm achieves a significantly higher speedup (739.6 on 2048 Cori KNL nodes) as compared to the vanilla synchronous SGD (276.6) while the model accuracy is almost not affected (<0.2% difference).

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Improving Computational Efficiency of Prognostics Algorithms in Resource-Constrained Settings

The field of prognostics and health management provides quantitative methods for monitoring and predicting the health of physical systems. Prognostics algorithms are useful in that they can be employed to assess the current state of a system, propagate the system state throughout time, and predict potential anomalies or failures that may occur. However, effective prognosis can be challenging to achieve in resource-constrained settings due to computational limitations and high computational latency, leading to obsolete predictions. Thus, computationally efficient and accurate algorithms are necessary for some prognostics applications. In this work, we implement three new algorithmic approaches to prediction (sampling methods, variable prediction time step, variable prediction sample size) with the goal of improving computational efficiency while minimizing decrease in model accuracy. To quantitatively analyze our results, we examine a use-case of degradation of a Lithium-ion battery. Notably, through this work it was found that none of the sampling approaches had a significant impact on computational efficiency or model accuracy in predicting EOD of the battery. However, our results show that prediction accuracy is highly dependent on the time step used, and that an appropriate time step can optimize both model accuracy and simulation efficiency. Finally, implementing a variable sample size also affected prediction, and our results show that tuning both the magnitude and timing of the sample size adjustment in an application-specific manner may prove useful in some applications. Taken together, our findings highlight the challenge of performing prognostics in resource-constrained settings, and illustrate the potential of developing new prediction algorithms to improve computational efficiency of prognosis.

Prognostics↗