Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “prediction methods”

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 OSTI2024

System and method for wave prediction

A method and system for prediction of wave properties include collecting time series data streams from one or more wave measurement devices and processing the data using a wave-prediction algorithm to identify the frequency components of the data and compute wave parameters. The wave-field is propagated in space and time to predict wave height, speed, and velocity at a target location. A sliding window approach is used to continuously update the prediction in real-time.

Previsic, Mirko↗
DOE OSTI2024

System and method for wave prediction

A method and system for prediction of wave properties include collecting time-series data streams from one or more wave measurement devices and processing the data to identify data parameters to establish boundary conditions of a numerical model. The numerical model may be used to compute a predicted wave field of time-series data for a variety of wave properties at a target location.

Previsic, Mirko↗
DOE OSTI2021

Texture Memory in Hexagonal Metals and Its Mechanism

Texture memory is a phenomenon in which retention of initial textures occurs after a complete cycle of forward and backward transformations, and it occurs in various phase-transforming materials including cubic and hexagonal metals such as steels and Ti and Zr alloys. Texture memory is known to be caused by the phenomena called variant selection, in which some of the allowed child orientations in an orientation relationship between the parent and child phases are preferentially selected. Without such variant selection, the phase transformations would randomize preferred orientations. In this article, the methods of prediction of texture memory and mechanisms of variant selections in hexagonal metals are explored. The prediction method using harmonic expansion of orientation distribution functions with the variant selection in which the Burgers orientation relationship, {110}β//{0001} α-hex <1$\bar{1}$1>β//$\langle2\bar{11}0\rangle$ α-hex , is held with two or more adjacent parent grains at the same time, called “double Burgers orientation relation (DBOR)”, is introduced. This method is shown to be a powerful tool by which to analyze texture memory and ultimately provide predictive capabilities for texture changes during phase transformations. Variation in nucleation and growth rates on special boundaries and an extensive growth of selected variants are also described. Analysis of textures of commercially pure Ti observed in situ by pulsed neutron diffraction reveals that the texture memory in CP-Ti is indeed quite well predicted by consideration of the mechanism of DBOR. The analysis also suggests that the nucleation and growth rates on the special boundary of 90° rotation about $\langle2\bar{11}0\rangle$ α-hex should be about three times larger than those of the other special boundaries, and the selected variants should grow extensively into not only one parent grain but also other grains in α-hex(hexagonal)→β(bcc) transformation. The model calculations of texture development during two consecutive cycles of α-hex→β→α-hex transformation in CP-Ti and Zr are also shown.

36 MATERIALS SCIENCE↗
DOE OSTI2023

A verification study of fatigue-based methods in API RP 1183 for estimating fatigue life of pipeline dents

Pipeline dents are mechanical damage caused by a third party in buried pipelines during construction or maintenance repair. Dents are recognized as a leading cause of pipeline failures for both liquid and gas transmission lines, and their assessment and management are critical to pipeline integrity. To meet the industry need, American Institute of Petroleum (API) in 2020 published a recommended practice (RP) 1183 – Assessment and Management of Pipeline Dents. This API code provides the pipeline industry “standard methods” for evaluating severity and fatigue life of pipeline dents with a single peak. This paper performed an initial verification study of fatigue-based methods prescribed in API RP 1183 for estimating fatigue life of pipeline dents. A brief review was first given to the dent fatigue-based methods, including three dent screening methods for estimating dent fatigue life in Sections 7.4.1, 7.4.2, and 7.4.3 of API 1183 and one dent assessment method for predicting fatigue life in Section 8.3.4. In order to verify these fatigue life prediction methods, a set of “standard examples” of dents provided in Annexes A.1 to A.5 of API RP 1183 were utilized in this study to calculate fatigue life, to determine restraint condition, and to evaluate both screening and assessment methods for single-peak dents. Through detailed calculations and comparisons, this study confirmed some specifications of API RP 1183, but also found some self-inconsistences and errors existing in this new API code. Furthermore, this includes the self-inconsistences between three screening methods and the self-inconsistence between the screening and assessment methods. On this basis, recommendations were made on how the operators should use API RP 1183, and what actions API should take to improve this code.

42 ENGINEERING↗
DOE OSTI2025

New Methods for Predicting Non-Born-Oppenheimer Chemistry

Current methods for modeling non-adiabatic molecular dynamics face fundamental limitations when treating geometric phase effects: quantum mechanical phenomena where nuclear wavepackets acquire phase shifts when encircling conical intersections. Existing approaches either neglect these effects entirely or rely on potential energy surfaces arising from the Born-Oppenheimer approximation, which introduce artificial singularities and can overestimate geometric phase contributions. This project developed a new theoretical framework based on exact factorization (XF) methods to overcome these limitations. We derived mathematical formulations for hybrid quantum-classical XF dynamics that selectively treat critical nuclear degrees of freedom quantum mechanically while propagating others classically. This approach addresses the computational intractability that has previously limited exact methods to toy systems. Key innovations include a new approach to systematically identifying nuclear coordinates requiring quantum treatment, as well as novel implementation strategies that interface with existing quantum chemistry codes. The project also developed a proof-of-concept code for treating Jahn-Teller systems and creation of educational materials on non-adiabatic dynamics geared at the graduate level. The theoretical framework developed will enable future systematically improvable calculations of nuclear quantum effects in realistic molecular systems, filling a critical gap in non-adiabatic dynamics methods. This foundation supports future development of predictive tools for designing energy-relevant photochemical processes where quantum coherence effects may be exploited to control reaction outcomes.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2023

Large Eddy Simulation of Gasoline Sprays in a Lagrangian–Eulerian Framework Using the High-Order Spectral Element Method

Predicting the spray evolution using simulations requires accurate modeling of the turbulent gas-phase flow field. Here, in this study, the high-order spectral-element method (SEM), implemented in the code Nek5000, was used to provide highly resolved solutions to the turbulent flow field. Spray modeling capabilities were implemented into the Nek5000 code. The spray is modeled in a Lagrangian–Eulerian (LE) framework, where the liquid is represented by discrete parcels of droplets. The method for coupling liquid and gas in the context of SEM is described, which allows for very fine meshes to be used without affecting the stability of the solution. Large-eddy simulations (LES) of the eight-hole ECN Spray G gasoline injector were conducted. Numerical results are compared against experimental data for liquid penetration, droplet size and gas velocity. The morphology of the multiplume spray is compared against experimental data. The effect of different spray injection inputs is analyzed. It was found that using a plume direction of 33 deg and an injection cone angle of 30 deg produced the best results overall. This work shows the applicability of SEM for spray modeling applications, where use of a high-order flow solver can help us understand the multiplume spray aerodynamics and how it leads to plume collapse under certain conditions. Results also highlight the need for tuning spray input parameters in the LE framework, even when high-fidelity gas flow solutions are possible.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2022

Large-eddy simulation of non-vaporizing sprays using the spectral-element method

Predictive simulations of high-pressure sprays require accurate representation of the turbulent gaseous flow field generated by liquid jet. Typically, the accuracy that can be obtained with low-order numerical methods (e.g. finite volume, finite element) is limited by stability issues in fine grids and the order of convergence of the method. In this work, we resolve the turbulent flow field in an Eulerian manner using the high-order spectral element method, coupled with a Lagrangian parcels approach to model the atomizing liquid jet. Large eddy simulations of single-hole sprays under non-evaporative conditions were conducted and compared against experimental data from Margot et al. (2008) and Spray A data from the Engine Combustion Network. The sensitivity of liquid penetration and droplet sizes to different breakup model parameters was studied. The effect of different numerical parameters, such as polynomial order of the solution (grid resolution), on liquid penetration was also analyzed. The method achieved grid-independent results using p-refinement, achieving finer resolution (by a factor of x 1.7 - x 3.5) in the gas-phase solution than in state-of-the-art simulations using the finite-volume method. Here the results showed good agreement with experimental data, demonstrating the ability of the current method to accurately capture liquid penetration and the shape of the spray.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Improving deep learning-based protein distance prediction in CASP14

Abstract Motivation Accurate prediction of residue–residue distances is important for protein structure prediction. We developed several protein distance predictors based on a deep learning distance prediction method and blindly tested them in the 14th Critical Assessment of Protein Structure Prediction (CASP14). The prediction method uses deep residual neural networks with the channel-wise attention mechanism to classify the distance between every two residues into multiple distance intervals. The input features for the deep learning method include co-evolutionary features as well as other sequence-based features derived from multiple sequence alignments (MSAs). Three alignment methods are used with multiple protein sequence/profile databases to generate MSAs for input feature generation. Based on different configurations and training strategies of the deep learning method, five MULTICOM distance predictors were created to participate in the CASP14 experiment. Results Benchmarked on 37 hard CASP14 domains, the best performing MULTICOM predictor is ranked 5th out of 30 automated CASP14 distance prediction servers in terms of precision of top L/5 long-range contact predictions [i.e. classifying distances between two residues into two categories: in contact (<8 Angstrom) and not in contact otherwise] and performs better than the best CASP13 distance prediction method. The best performing MULTICOM predictor is also ranked 6th among automated server predictors in classifying inter-residue distances into 10 distance intervals defined by CASP14 according to the precision of distance classification. The results show that the quality and depth of MSAs depend on alignment methods and sequence databases and have a significant impact on the accuracy of distance prediction. Using larger training datasets and multiple complementary features improves prediction accuracy. However, the number of effective sequences in MSAs is only a weak indicator of the quality of MSAs and the accuracy of predicted distance maps. In contrast, there is a strong correlation between the accuracy of contact/distance predictions and the average probability of the predicted contacts, which can therefore be more effectively used to estimate the confidence of distance predictions and select predicted distance maps. Availability and implementation The software package, source code and data of DeepDist2 are freely available at https://github.com/multicom-toolbox/deepdist and https://zenodo.org/record/4712084#.YIIM13VKhQM. Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

A tutorial review of machine learning-based model predictive control methods

Abstract This tutorial review provides a comprehensive overview of machine learning (ML)-based model predictive control (MPC) methods, covering both theoretical and practical aspects. It provides a theoretical analysis of closed-loop stability based on the generalization error of ML models and addresses practical challenges such as data scarcity, data quality, the curse of dimensionality, model uncertainty, computational efficiency, and safety from both modeling and control perspectives. The application of these methods is demonstrated using a nonlinear chemical process example, with open-source code available on GitHub. The paper concludes with a discussion on future research directions in ML-based MPC.

Wu, Zhe [Department of Chemical and Biomolecular E↗
DOE OSTI2023

System and method for predicting the presence of rare earth elements

A system for predicting rare earth elements (REEs) in a feedstock sample includes a measurement instrument that records a measurement for a sample, a processor communicatively coupled to the measuring instrument, and a memory communicatively coupled to the processor and containing machine readable instructions that, when executed by the processor, cause the processor to correlate the measurement series using a model; and predict a presence of one or more rare earth element based at least in part on the correlation. A method for predicting rare earth elements includes measuring feedstock samples via XRF or PGNAA, to generate a measurements of elements of interest with a lower atomic weight than REEs; correlating the measurements with a model; and predicting a presence of one or more rare earth elements based at least in part on the correlation.

Fuka, Matthew↗
DOE OSTI2023

3D-equivariant graph neural networks for protein model quality assessment

Quality assessment (QA) of predicted protein tertiary structure models plays an important role in ranking and using them. With the recent development of deep learning end-to-end protein structure prediction techniques for generating highly confident tertiary structures for most proteins, it is important to explore corresponding QA strategies to evaluate and select the structural models predicted by them since these models have better quality and different properties than the models predicted by traditional tertiary structure prediction methods. We develop EnQA, a novel graph-based 3D-equivariant neural network method that is equivariant to rotation and translation of 3D objects to estimate the accuracy of protein structural models by leveraging the structural features acquired from the state-of-the-art tertiary structure prediction method—AlphaFold2. We train and test the method on both traditional model datasets (e.g. the datasets of the Critical Assessment of Techniques for Protein Structure Prediction) and a new dataset of high-quality structural models predicted only by AlphaFold2 for the proteins whose experimental structures were released recently. Our approach achieves state-of-the-art performance on protein structural models predicted by both traditional protein structure prediction methods and the latest end-to-end deep learning method—AlphaFold2. It performs even better than the model QA scores provided by AlphaFold2 itself. The results illustrate that the 3D-equivariant graph neural network is a promising approach to the evaluation of protein structural models. Integrating AlphaFold2 features with other complementary sequence and structural features is important for improving protein model QA.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

A Comparative Study of Layer Heating and Continuous Heating Methods on Prediction Accuracy of Residual Stresses in Selective Laser Melted Tube Samples

Thermal distortion and residual stresses are two important factors that affect the quality and reliability of steel parts manufactured by laser powder bed fusion (LPBF) processes. A cost-effective model for evaluation of those heat effects is needed to refine the manufacturing process and provides insights into the product design and heat treatment. In this study, the layer heating method and sophisticated track-layer scanning method were applied to simulate the thermo-mechanical response of IN625 tube parts built by LPBF. Based on the similarity of temperature field in each layer deposit, a swept mesh was constructed to perform the thermal analysis for top layer, with the rest of layers referring to the temperature by node number offsetting. A novel explicit finite element analysis code accelerated by graphics processing unit was used for the massive-element numerical analysis. The computational accuracy and efficiency of the layer heating and track-layer scanning methods were compared in detail. It is shown that layer heating method can efficiently capture the pattern of stress distribution with reasonable accuracy in stress magnitude. The grouped track-layer scanning method can predict the residual stress and strain more accurately at a higher cost (5 ~ 10×). The elastic strain distribution was compared with the measurement by X-ray diffraction, confirming the accuracy of residual stress prediction.

36 MATERIALS SCIENCE↗
DOE OSTI2021

A prediction interval method for uncertainty quantification of regression models

This paper considers calculation of prediction intervals (PIs) by neural networks (NNs) for quantifying uncertainty in regression tasks, so as to provide fast, accurate and robust emulators to accelerate scientific simulations. We propose a novel method to learn lower and upper bounds of the PI using independent NNs without defining an exclusive loss. Our method requires no distributional assumption, does not introduce extra hyper-parameters, and can effectively identify out-of-distribution samples and quantify their uncertainty. We demonstrate advantages of our method using a benchmark problem and two real-world scientific applications.

Zhang, Pei↗
DOE OSTI2021

A New Approach for Calculating the Alpha-Decay Half-Life for the Heavy and Super-heavy Elements and an Exact A Priori Result for Beyllium-8

A new general method for calculating the alpha decay half-life is presented. The method predicts an a priori exact value for the beryllium-8 half-life. Beryllium-8 is an exception to the current alpha decay theory captured in the Geiger-Nuttal law. The new method predicts the beryllium-8 alpha half-life using only constants and measured isotopic mass. The method also reliably predicts all the heavier isotope alpha decay half-lives consistent with the Geiger-Nuttal law. With respect to current theory, the inability of the Geiger-Nuttal Law to predict the alpha-decay half-life in the case of beryllium-8 has led to consideration of other decay mechanisms for this isotope, such as fission for example. One result is that given the consistency of the new method presented here for all isotopes including an exact a priori result for beryllium-8, the evidence strongly suggests that the beryllium-8 decay is in fact an alpha decay. A second result is that the method definitively demonstrates that the entire rest mass of the two helium-4 electrons is converted to energy in the decay process and this energy becomes part of the emitted alpha particle kinetic energy.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2022

Machine Learning Assisted HPC Workload Trace Generation for Leadership Scale Storage Systems

Monitoring and analyzing a wide range of I/O activities in an HPC cluster is important in maintaining mission-critical performance in a large-scale, multi-user, parallel storage system. Center-wide I/O traces can provide high-level information and fine-grained activities per application or per user running in the system. Studying such large-scale traces can provide helpful insights into the system. It can be used to develop predictive methods for making predictive decisions, adjusting scheduling policies, or providing decisions for the design of next-generation systems. However, sharing real-world I/O traces to expedite such research efforts leaves a few concerns; i) the cost of sharing the large traces is expensive due to this large size, and ii) privacy concern is an issue.We address such issues by building an end-to-end machine learn- ing (ML) workflow that can generate I/O traces for large-scale HPC applications. We leverage ML based feature selection and gener- ative models for I/O trace generation. The generative models are trained on I/O traces collected by the darshan I/O characterization tool over a period of one year. We present a two-step generation process consisting of two deep-learning models, called the feature generator and the trace generator. The combination of two-step generative models provides robustness by reducing the bias of the model and accounting for the stochastic nature of the I/O traces across different runs of an application. We evaluate the performance of the generative models and show that the two-step model can generate time-series I/O traces with less than 20% root mean square error.

Paul, Arnab↗
DOE OSTI2024

A Simplified Method for Predicting Shaker Voltage in IMMATs

Impedance Matched Multi-Axis Tests (IMMATs) can replicate in-service vibration induced stress more accurately than single axis shaker table tests as they can better match a part’s operational boundary conditions and excite it in multiple degrees of freedom simultaneously. Here, the shakers used in IMMATs are less powerful than shaker tables, so shaker force limits can be exceeded during tests if they are not placed adequately for the desired environment. The ability to predict shaker voltage and force before performing a test is, therefore, helpful in selecting shaker locations so that their limits are not exceeded. In this study, electrodynamic shakers were modeled as discrete electromechanical systems, and the shaker parameters were chosen to match experimentally obtained acceleration/voltage frequency response functions (FRFs). These models were coupled to a finite element model of the device under test (DUT) via dynamic substructuring, and the substructured model was demonstrated to accurately predict shaker voltage as well as the error in reproducing the environment at multiple accelerometer locations. A simple method called the FRF Multiplication method, in which the FRF of the substructured system is approximated as the product of two separate FRFs of the shaker and DUT respectively, was proposed and applied to the same system, yielding similar voltage and error predictions to those obtained using substructuring. Simple case studies were presented to explore the applicability of the proposed method, and it was demonstrated to have similar accuracy to the substructuring method in a range of cases. Additionally, we showed that while it was not possible to derive a unique model of the shakers from acceleration/voltage FRFs alone, the models that could be obtained were sufficient to predict test error almost perfectly and shaker voltage with less than 40 percent error.

42 ENGINEERING↗
DOE OSTI2021

Data-Driven Forgetting and Discount Factors for Vehicle Speed Forecasting in Ecological Adaptive Cruise Control

Abstract This paper investigates temporal correlations in human driving behavior using real-world driving to improve speed forecasting accuracy. These correlations can point to a measurement weighting function with two parameters: a forgetting factor for past speed measurements that the vehicle itself drove with, and a discount factor for the speeds of vehicles ahead based on information from vehicle-to-vehicle communication. The developed weighting approach is applied to a vehicle speed predictor using polynomial regression, a prediction method well-known in the literature. The performance of the developed approach is then assessed in both real-world and simulated traffic scenarios for accuracy and robustness. The new weighting method is applied to an ecological adaptive cruise control system, and its influence is analyzed on the prediction accuracy and the performance of the ecological adaptive cruise control in an electric vehicle powertrain model. The results show that the new prediction method improves energy saving from the eco-driving by up to 4.7% compared to a baseline least-square-based polynomial regression. This is a 10% improvement over the constant speed/acceleration model, a conventional speed predictor.

Automation & Control Systems↗
DOE OSTI2022

A method to predict texture effect on ion beam channeling analysis of polycrystals and the application to study the mosaic spreading effect in highly oriented pyrolytic graphite

We propose a method to convert the channeling Rutherford backscattering spectrum yield map of a single crystal to a polycrystal through a matrix rotation technique. The rotation matrix is determined by the deviation of the crystal axial direction from the original z axis. The final yield map is created after averaging the rotated yields using the texture function as the weight factor. For highly oriented pyrolytic graphite (HOPG) exhibiting mosaic spread, the method leads to a Gaussian kernel averaging of the map obtained from a single crystal. The yield map of a single crystal is obtained by a simulation of ion trajectories in a potential field described by Moliere screened Coulomb potentials. Yield maps are calculated under various σ values (standard deviations of mosaic spread). The simulated results are compared with experimental results obtained using 1.2 MeV alpha particle. σ is extracted through the best fitting, demonstrating that the method can be used to obtain texture details. The effects of mosaic spread on minimum yield χ min and the half-width at half maximum of angular scans ψ ½ are systematically modeled and compared with previous theoretical equations. The study also shows that previous theoretical equations are valid only at small σ values. The proposed method can be applied to any type of polycrystal and is not limited to HOPG. It provides near-surface mosaic spread and crystallography information with a longitudinal depth resolution of tens of nanometers and is not influenced by grain shapes.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗