Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “factorization machine”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

SU-E-T-631: Preliminary Results for Analytical Investigation Into Effects of ArcCHECK Setup Errors

Purpose: As three-dimensional diode arrays increase in popularity for patient-specific quality assurance for intensity-modulated radiation therapy (IMRT), it is important to evaluate an array’s susceptibility to setup errors. The ArcCHECK phantom is set up by manually aligning its outside marks with the linear accelerator’s lasers and light-field. If done correctly, this aligns the ArcCHECK cylinder’s central axis (CAX) with the linear accelerator’s axis of rotation. However, this process is prone to error. This project has developed an analytical expression including a perturbation factor to quantify the effect of shifts. Methods: The ArcCHECK is set up by aligning its machine marks with either the sagittal room lasers or the light-field of the linear accelerator at gantry zero (IEC). ArcCHECK has sixty-six evenly-spaced SunPoint diodes aligned radially in a ring 14.4 cm from CAX. The detector response function (DRF) was measured and combined with inverse-square correction to develop an analytical expression for output. The output was calculated using shifts of 0 (perfect alignment), +/−1, +/−2 and +/−5 mm. The effect on a series of simple inputs was determined: unity, 1-D ramp, steps, and hat-function to represent uniform field, wedge, evenly-spaced modulation, and single sharp modulation, respectively. Results: Geometric expressions were developed with perturbation factor included to represent shifts. DRF was modeled using sixth-degree polynomials with correlation coefficient 0.9997. The output was calculated using simple inputs such as unity, 1-D ramp, steps, and hat-function, with perturbation factors of: 0, +/−1, +/−2 and +/−5 mm. Discrepancies have been observed, but large fluctuations have been somewhat mitigated by aliasing arising from discrete diode placement. Conclusion: An analytical expression with perturbation factors was developed to estimate the impact of setup errors on an ArcCHECK phantom. Presently, this has been applied to simple functions, but is being developed to handle more realistic clinical cases.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2023

Correcting distortions of thin-walled machined parts by machine hammer peening

Thin-walled aerostructural components frequently get distorted after the machining process. Reworking to correct distortions or eventually rejecting parts significantly increases the cost. This paper proposes a new approach to correct distortions in thin-walled components by strategically applying hammer peening on target surfaces of a machined component. Aluminium alloy 7475-T7351 was chosen for this research. The study was divided in two stages. First, the residual stresses (RS) induced by four different pneumatic hammer peening conditions (modifying the stepover distance and initial offset) were characterised in a test coupon, and one of the conditions was selected for the next stage. In the second stage, a FEM model was used to predict distortions caused by machining in a representative workpiece. Then, the RS induced by hammer peening were included in an FEM model to define two hammer peening strategies (varying the coverage area) to analyse the capability to reduce distortions. Two workpieces were machined and then treated with the simulated hammer peening strategies for experimental validation. Results in the test coupon showed that pneumatic hammer peening can generate high compressive RS (-50 to –350 MPa) up to 800 μm depth, with their magnitude increasing with a reduced stepover distance. Application of hammer peening over 4% of the surface of the representative workpiece reduced the machining-induced distortions by 37%, and a coverage area of 100% led to and overcorrection by a factor of five. This confirms that hammer peening can be strategically applied (in target areas and changing the percentage of coverage) to correct low or severe distortions.

42 ENGINEERING↗
DOE OSTI2016

TH-AB-201-09 [Medical Physics, Jun 2016, v. 43(6)]

Purpose: To determine the k-(Q-msr,Q)^(f-msr,f-ref ) factor, introduced in the small field formalism for five common type chambers used in the calibration of Leksell Gamma-Knife Perfexion model over a range of different phantom electron densities. Methods: Five chamber types including Exradin-A16, A14SL, A14, A1SL and IBA-CC04 are modeled in EGSnrc and PENELOPE Monte Carlo codes using the blueprints provided by the manufacturers. The chambers are placed in a previously proposed water-filled phantom and four 16-cm diameter spherical phantoms made of liquid water, Solid Water, ABS and polystyrene. Dose to the cavity of the chambers and a small water volume are calculated using EGSnrc/PENELOPE codes. The calculations are performed over a range of phantom electron densities for two chamber orientations. Using the calculated dose-ratio in reference and machine specific reference field, the k-(Q-msr,Q)^(f-msr,f-ref ) factor can be determined. Results: When chambers are placed along the symmetry axis of the collimator block (z-axis), the CC04 requires the smallest correction followed by A1SL and A16. However, when detectors are placed perpendicular to z-axis, A14SL needs the smallest and A16 the largest correction. Moreover, an increase in the phantom electron density results in a linear increase in the k-(Q-msr,Q)^(f-msr,f-ref ). Depending on the chambers, the agreement between this study and a previous study performed varies between 0.05–0.70% for liquid water, 0.07–0.85% for Solid Water and 0.00–0.60% for ABS phantoms. After applying the EGSnrc-calculated k-(Q-msr,Q)^(f-msr,f-ref ) factors for A16 to the previously measured dose-rates in liquid water, Solid Water and ABS normalized to the dose-rate measured with TG-21 protocol and ABS phantom, the dose-rate ratios are found to be 1.004±0.002, 0.996±0.002 and 0.998±0.002 (3σ) respectively. Conclusion: Knowing the electron density of the phantoms, the calculated k-(Q-msr,Q)^(f-msr,f-ref ) values in this work will enable users to apply the appropriate correction for their own specific phantom material. LM acknowledges partial support by the CREATE Medical Physics Research Training Network grant of the Natural Sciences and Engineering Research Council (Grant number: 432290)

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Controls and variability of soil respiration temperature sensitivity across China

Understanding the temperature sensitivity (Q 10 ) of soil respiration is critical for benchmarking the potential intensity of regional and global terrestrial soil carbon fluxes-climate feedbacks. Although field observations have demonstrated the strong spatial heterogeneity of Q 10 , a significant knowledge gap still exists regarding to the factors driving spatial and temporal variabilities of Q 10 at regional scales. Here, therefore, we used a machine learning approach to predict Q 10 from 1994 to 2016 with a spatial resolution of 1 km across China from 515 field observations at 5 cm soil depth using climate, soil and vegetation variables. Predicted Q 10 varied from 1.54 to 4.17, with an area-weighted average of 2.52. There was no significant temporal trend for Q 10 (p = 0.32), but annual vegetation production (indicated by normalized difference vegetation index, NDVI) was positively correlated to it (p < 0.01). Spatially, soil organic carbon (SOC) was the most important driving factor in 62 % of the land area across China, and varied greatly, demonstrating soil controls on the spatial pattern of Q 10 . These findings highlighted different environmental controls on the spatial and temporal pattern of soil respiration Q 10 , which should be considered to improve global biogeochemical models used to predict the spatial and temporal patterns of soil carbon fluxes to ongoing climate change.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2017

Optimization of human factors methods within the design cycle - 178

Amec Foster Wheeler developed and implemented the Human Factors Engineering program to analyze the design of a large-scale electrical component readying for installation in a nuclear facility. The HFE program applied the program elements from NUREG-0711 throughout the detailed design phase of the project. This paper provides: (1) a review of the Human Factors Engineering (HFE) program methodology that was implemented, and (2) a summary of the integrated approach to the design, verification, and validation of the Human-Machine Interface (HMI) display and mimic panel. In general, it was found that a collaborative approach between the design team, equipment vendor, HFE, and the end-users of the component created a more streamlined approach to the design and allowed for user feedback to be implemented throughout the progression of the design. However, the trade-off to this approach is that an integrated validation is likely required once all design aspects are complete, to ensure the modification is acceptable in the context of the overall system. (authors)

42 ENGINEERING↗
DOE OSTI2023

A review on recent machine learning applications for imaging mass spectrometry studies

Imaging mass spectrometry (IMS) is a powerful analytical technique widely used in biology, chemistry, and materials science fields that continue to expand. IMS provides a qualitative compositional analysis and spatial mapping with high chemical specificity. The spatial mapping information can be 2D or 3D depending on the analysis technique employed. Due to the combination of complex mass spectra coupled with spatial information, large high-dimensional datasets (hyperspectral) are often produced. Therefore, the use of automated computational methods for an exploratory analysis is highly beneficial. The fast-paced development of artificial intelligence (AI) and machine learning (ML) tools has received significant attention in recent years. These tools, in principle, can enable the unification of data collection and analysis into a single pipeline to make sampling and analysis decisions on the go. There are various ML approaches that have been applied to IMS data over the last decade. Here, in this review, we discuss recent examples of the common unsupervised (principal component analysis, non-negative matrix factorization, k-means clustering, uniform manifold approximation and projection), supervised (random forest, logistic regression, XGboost, support vector machine), and other methods applied to various IMS datasets in the past five years. The information from this review will be useful for specialists from both IMS and ML fields since it summarizes current and representative studies of computational ML-based exploratory methods for IMS.

47 OTHER INSTRUMENTATION↗
DOE OSTI2014

Optimization of the Shielding of a Canister Loading Machine with the deterministic Code Attila{sup R} - Paper 61

In general shielding components are designed with very conservative assumptions and boundary conditions to cover all possibly occurring situations. This often results in an overestimated shielding design in a lot of different fields of application; the shielding of components will be heavier and more expensive than needed. In most cases an initial shielding calculation in the design process shows an unbalanced shielding effect of the shielded radiation source. Sometimes the dose rate limits are slightly exceeded in several areas. From this arises a need for an optimization of the shielding structure. The goal is to decrease the radiation level in the accessible areas to keep the dose rate limits and to reduce the amount of the used shielding material if possible. Shielding material is moved from areas with higher shielding factors to which with lower ones. The optimization process is shown in an example of a loading machine for vitrified waste in canisters. The design review with an optimization of the shielding geometry was done with the deterministic code Attila{sup R} of the company Transpire, Inc. Attila{sup R} is a powerful tool to calculate and optimize the shielding effect of handling facilities for radioactive materials. It is easy to handle with its CAD integration module and an easy setup of the calculation cases. Areas with a low shielding effect which results in an exceeding of the dose rate limits can be found quickly with the illustration due to an Iso-Dose-Rate-Surface in the 3D Model. In these areas the shielding effect has to be improved. Otherwise there is material, which has no significant influence on the shielding effect and makes the handling facilities heavier and more expensive as needed. The optimization of the shielding design of the canister loading machine in the current case results in lower dose rates outside the facility to keep the dose rate limits. The demand of shielding material could be reduced and resulted in a cost reduction on the facility itself and furthermore in the caring facilities to transport this facility (cranes, etc.). (authors)

42 ENGINEERING↗
DOE OSTI2015

Collective I/O Tuning Using Analytical and Machine-Learning Models

The ever larger demand of scientific applications for computation and data is currently driving a continuous increase in scale of parallel computers. The inherent complexity of scaling up a computing systems in terms of both hardware and software stack exposes an increasing number of factors impacting the performance and complicating the process of optimization. In particular, the optimization of parallel I/O has become increasingly challenging due to increasing storage hierarchy and well known performance variability of shared storage systems. This paper focuses on model-based autotuning of the two-phase collective I/O algorithm from a popular MPI distribution on the Blue Gene/Q architecture. We propose a novel hybrid model, constructed as a composition of analytical models for communication and storage operations and black-box models for the performance of the individual operations. We perform an in-depth study of the complexity involved in performance modeling including architecture, software stack and noise. In particular we address this challenges of modeling the performance of shared storage systems by building a benchmark that helps synthesizing factors such as topology, file caching, and noise. The experimental results show that the hybrid approach produces significantly better results than state-of-the-art machine learning approaches and shows a higher robustness to noise, at the cost of a higher modeling complexity

I/O performance modeling↗
DOE OSTI2022

Construction of Women’s All-Around Speed Skating Event Performance Prediction Model and Competition Strategy Analysis Based on Machine Learning Algorithms

Introduction Accurately predicting the competitive performance of elite athletes is an essential prerequisite for formulating competitive strategies. Women’s all-around speed skating event consists of four individual subevents, and the competition system is complex and challenging to make accurate predictions on their performance. Objective The present study aims to explore the feasibility and effectiveness of machine learning algorithms for predicting the performance of women’s all-around speed skating event and provide effective training and competition strategies. Methods The data, consisting of 16 seasons of world-class women’s all-around speed skating competition results, used in the present study came from the International Skating Union (ISU). According to the competition rules, distinct features are filtered using lasso regression, and a 5,000 m race model and a medal model are built using a fivefold cross-validation method. Results The results showed that the support vector machine model was the most stable among the 5,000 m race and the medal models, with the highest AUC (0.86, 0.81, respectively). Furthermore, 3,000 m points are the main characteristic factors that decide whether an athlete can qualify for the final. The 11th lap of the 5,000 m, the second lap of the 500 m, and the fourth lap of the 1,500 m are the main characteristic factors that affect the athlete’s ability to win medals. Conclusion Compared with logistic regression, random forest, K-nearest neighbor, naive Bayes, neural network, support vector machine is a more viable algorithm to establish the performance prediction model of women’s all-around speed skating event; excellent performance in the 3,000 m event can facilitate athletes to advance to the final, and athletes with outstanding performance in the 500 m event are more likely competitive for medals.

Liu, Meng↗
DOE OSTI2019

A theory of quark vs. gluon discrimination

Understanding jets initiated by quarks and gluons is of fundamental importance in collider physics. Efficient and robust techniques for quark versus gluon jet discrimination have consequences for new physics searches, precision αs studies, parton distribution function extractions, and many other applications. Numerous machine learning analyses have attacked the problem, demonstrating that good performance can be obtained but generally not providing an understanding for what properties of the jets are responsible for that separation power. In this paper, we provide an extensive and detailed analysis of quark versus gluon discrimination from first-principles theoretical calculations. Working in the strongly-ordered soft and collinear limits, we calculate probability distributions for fixed N-body kinematics within jets with up through three resolved emissions (O(α$3\atop{s}$)). This enables explicit calculation of quantities central to machine learning such as the likelihood ratio, the area under the receiver operating characteristic curve, and reducibility factors within a well-defined approximation scheme. Further, we relate the existence of a consistent power counting procedure for discrimination to ideas for operational flavor definitions, and we use this relationship to construct a power counting for quark versus gluon discrimination as an expansion in e C F -C A $\ll$1, the exponential of the fundamental and adjoint Casimirs. Our calculations provide insight into the discrimination performance of particle multiplicity and show how observables sensitive to all emissions in a jet are optimal. We compare our predictions to the performance of individual observables and neural networks with parton shower event generators, validating that our predictions describe the features identified by machine learning

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Employing Machine Learning for New Particle Formation Identification and Mechanistic Analysis: Insights From a Six‐Year Observational Study in the Southern Great Plains

We present a supervised machine learning (ML) framework to automatically identify new particle formation (NPF) events and analyze key atmospheric factors associated with their occurrence and growth. We applied ML to detect NPF events using start time and particle concentrations across size ranges, while identifying atmospheric variables including ambient temperature, relative humidity, solar radiation intensity (SRI), wind speed, wind direction, boundary layer height, total organics, sulfate, nitrate, total surface area concentration, sulfur dioxide, and turbulent kinetic energy (TKE). We analyzed a 6-year data set from the Atmospheric Radiation Measurement at the Southern Great Plains (SGP) site in Oklahoma, USA. Using long-term ground-based measurements, we identified NPF events and applied Random Forest Classifiers, which achieved 90%–95% prediction accuracy. Feature importance analysis highlighted SRI, relative humidity, and ambient temperature as the most influential variables, contributing normalized importances of 28%, 17%, and 10%. Partial Dependence Plots (PDPs) indicated that higher SRI and lower relative humidity were critical in promoting NPF formation at SGP. Seasonally, NPF events were more frequent in winter (42.1%) and spring (35.5%), and least in summer (4.0%). Particle growth rates also exhibited a seasonal variation, with the lowest in winter (below 2 nm hr −1 ) and highest in late spring and early summer (exceeding 5 nm hr −1 ). Temperature, turbulent kinetic energy, and aerosol properties were the primary factors of growth rate variability. This study advances predictive modeling of NPF, offers insights for future campaign deployments, and demonstrates the effectiveness of ML in understanding the formation and growth of atmospheric aerosols.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Evaluating county-level lung cancer incidence from environmental radiation exposure, PM 2.5 , and other exposures with regression and machine learning models

Characterizing the interplay between exposures shaping the human exposome is vital for uncovering the etiology of complex diseases. For example, cancer risk is modified by a range of multifactorial external environmental exposures. Environmental, socioeconomic, and lifestyle factors all shape lung cancer risk. However, epidemiological studies of radon aimed at identifying populations at high risk for lung cancer often fail to consider multiple exposures simultaneously. For example, moderating factors, such as PM 2.5 , may affect the transport of radon progeny to lung tissue. This ecological analysis leveraged a population-level dataset from the National Cancer Institute’s Surveillance, Epidemiology, and End-Results data (2013–17) to simultaneously investigate the effect of multiple sources of low-dose radiation (gross γ activity and indoor radon) and PM 2.5 on lung cancer incidence rates in the USA. County-level factors (environmental, sociodemographic, lifestyle) were controlled for, and Poisson regression and random forest models were used to assess the association between radon exposure and lung and bronchus cancer incidence rates. Tree-based machine learning (ML) method perform better than traditional regression: Poisson regression: 6.29/7.13 (mean absolute percentage error, MAPE), 12.70/12.77 (root mean square error, RMSE); Poisson random forest regression: 1.22/1.16 (MAPE), 8.01/8.15 (RMSE). The effect of PM 2.5 increased with the concentration of environmental radon, thereby confirming findings from previous studies that investigated the possible synergistic effect of radon and PM 2.5 on health outcomes. In summary, the results demonstrated (1) a need to consider multiple environmental exposures when assessing radon exposure’s association with lung cancer risk, thereby highlighting (1) the importance of an exposomics framework and (2) that employing ML models may capture the complex interplay between environmental exposures and health, as in the case of indoor radon exposure and lung cancer incidence.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗
DOE OSTI2020

Ambient Pressure XPS Investigation of Thermally Stable Halide Perovskite Solar Cells via Post-Treatment

Long-term thermal stability is one limiting factor that impedes the commercialization of perovskite solar cell. Inspired by our prior results from machine learning, we discover that coating a thin layer of 4,4’-dibromotriphenylamine (DBTPA) on top of a CH 3 NH 3 PbI 3 (MAPbI 3 ) thin film can improve the stability of the resultant perovskite solar cells. The passivated perovskite solar cells retained more than 96% of their initial power conversion efficiency over 1000 h at 85 °C in N2 atmosphere without encapsulation. Near-ambient pressure x-ray photoelectron spectroscopy (NAP-XPS) was employed to investigate the evolution of composition and evaluate thermal and moisture stability by in-situ studies. A comparison between pristine MAPbI 3 films and DBTPA-treated films shows the DBTPA treatment suppresses the escape of iodide and methylamine up to 150 °C under 5 mbar humidity. Furthermore, we have used ATR-FTIR and XPS to probe the interactions between DBTPA and MAPbI3 surface. The results prove that DBTPA coordinate with perovskite by Lewis acid-base and cation-π interaction. Compared with the 19.9% efficiency of pristine sample, the champion efficiency of passivated sample reaches 20.6%. Finally, our results reveal DBTPA as a new post-treating molecule that not only leads to the improvement of photovoltaic efficiency but also thermal and moisture stability.

14 SOLAR ENERGY↗
DOE OSTI2025

Resolving the dynamic correlated disorder in KTa 1- x Nb x O 3

Understanding the complex temporal and spatial correlations of ions in disordered perovskite oxides is critical to rationalize their functional properties. Here, in this study, we provide new insights into the longstanding controversy regarding the off-centering of transition metal ions in the archetypal ferroelectric alloy KTa 1-x Nb x O 3 (KTN). By mapping the full energy (E) and wavevector (Q) dependence of the dynamical structure factor S(Q, E) using neutron scattering, and rationalizing our observations with atomistic simulations leveraging machine learning, we fully resolve the static vs dynamic nature of diffuse scattering sheets, as well as their composition (x) and temperature dependence. Our first-principles simulations, extended with machine-learning molecular dynamics, reproduce both inelastic neutron spectra and diffuse features, and establish how dynamically-correlated transition metal off-centerings couple to phonons, unifying local and collective viewpoints. This study sheds new light into an exemplary ferroelectric systems and shows the importance of mapping the full S(Q, E) to reveal critical spatio-temporal correlations of atomic disorder from which functional properties emerge.

42 ENGINEERING↗
DOE OSTI2024

Automatic Loss Factor Modeling and Attribution on Unlabeled PV Energy Data

We present a novel approach for modeling the loss factors of photovoltaic power generation systems (PV systems). This method is a white-box machine learning model built on convex optimization that is fast, interpretable, and auditable. It takes as an input the measured daily energy produced by the system, over a multi-year period, and returns a multiplicative decomposition model of the daily energy signal and full attribution of the total energy loss to each feature. The methods section of this paper has two major components: (1) the description of the signal decomposition (SD) model, expressed in the SD framework, and (2) the attribution of total energy losses via Shapley values. We validate the method on synthetic and open-source data sets and compare to similar methods from the literature.

artificial intelligence↗
DOE OSTI2026

Quantum-inspired weight-constrained neural network: Reducing variable numbers by 100× compared to standard neural networks

Although quantum machine learning has shown great promise, the practical application of quantum computers remains constrained in the noisy intermediate-scale quantum era. To take advantage of quantum machine learning, we investigate the underlying mathematical principles of these quantum models and find that the quantum neural network with amplitude encoding is equivalent to a weight-constrained neural network. Motivated by this discovery, we develop a classical weight-constrained neural network. We find that this approach can reduce the number of variables in a classical neural network by a factor of 135 while preserving its accuracy. In addition, we develop a dropout method to enhance the robustness of quantum machine learning models, which are highly susceptible to adversarial attacks. This technique can also be applied to improve the adversarial robustness of the classical weight-constrained neural network, which is essential for industry applications, such as self-driving vehicles. Our work offers an approach to reduce the complexity of large classical neural networks, addressing a critical challenge in machine learning.

quantum algorithms & computation↗
DOE OSTI2024

Machine Learning-Driven Reliability Estimation of PV Inverters Considering Alert-Ambient Variability

Weather-induced spatio-temporal degradation limits outdoor PV inverter lifetime and reliability, necessitating advanced data analysis. This study employs a top-down, data-driven approach utilizing multiple machine learning (ML) algorithms to estimate inverter reliability in a 1.4 MW PV power plant, considering factors such as irradiance, humidity, temperature, time of day, and weather conditions. An extensive alert dataset from 17 identical inverters, including alert types, propagation, and frequency, reveals significant correlations with environmental factors and inverter output power, enabling the construction of a performance reliability model. Dual-stage supervised-ML models are evaluated for accuracy, with the ‘classification-regression’ model by an artificial neural network (ANN) tested on the averaged “Alert-Ambient” dataset, which is outperformed by ‘clustering-regression’ models using random forest (RF) and K-Nearest Neighbors (KNN) on individual inverter datasets. K-means clustering applies principal component analysis to reduce dimensions, achieving improved accuracy beyond the 80% achieved by ANN on the averaged dataset. Second-stage regression estimates inverter reliability with a mean square error of 0.0195 on the averaged dataset and as low as 0.002 on individual inverter datasets using RF. Furthermore, these findings highlight the method's suitability for estimating PV inverter output reliability under ambient conditions, essential for digital twin development and related applications.

14 SOLAR ENERGY↗