Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “performance modeling and prediction”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Improving protein tertiary structure prediction by deep learning and distance prediction in CASP14

Abstract Substantial progresses in protein structure prediction have been made by utilizing deep‐learning and residue‐residue distance prediction since CASP13. Inspired by the advances, we improve our CASP14 MULTICOM protein structure prediction system by incorporating three new components: (a) a new deep learning‐based protein inter‐residue distance predictor to improve template‐free (ab initio) tertiary structure prediction, (b) an enhanced template‐based tertiary structure prediction method, and (c) distance‐based model quality assessment methods empowered by deep learning. In the 2020 CASP14 experiment, MULTICOM predictor was ranked seventh out of 146 predictors in tertiary structure prediction and ranked third out of 136 predictors in inter‐domain structure prediction. The results demonstrate that the template‐free modeling based on deep learning and residue‐residue distance prediction can predict the correct topology for almost all template‐based modeling targets and a majority of hard targets (template‐free targets or targets whose templates cannot be recognized), which is a significant improvement over the CASP13 MULTICOM predictor. Moreover, the template‐free modeling performs better than the template‐based modeling on not only hard targets but also the targets that have homologous templates. The performance of the template‐free modeling largely depends on the accuracy of distance prediction closely related to the quality of multiple sequence alignments. The structural model quality assessment works well on targets for which enough good models can be predicted, but it may perform poorly when only a few good models are predicted for a hard target and the distribution of model quality scores is highly skewed. MULTICOM is available at https://github.com/jianlin-cheng/MULTICOM_Human_CASP14/tree/CASP14_DeepRank3 and https://github.com/multicom-toolbox/multicom/tree/multicom_v2.0 .

59 BASIC BIOLOGICAL SCIENCES↗
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 OSTI2020

On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events

Abstract Machine learned force fields typically require manual construction of training sets consisting of thousands of first principles calculations, which can result in low training efficiency and unpredictable errors when applied to structures not represented in the training set of the model. This severely limits the practical application of these models in systems with dynamics governed by important rare events, such as chemical reactions and diffusion. We present an adaptive Bayesian inference method for automating the training of interpretable, low-dimensional, and multi-element interatomic force fields using structures drawn on the fly from molecular dynamics simulations. Within an active learning framework, the internal uncertainty of a Gaussian process regression model is used to decide whether to accept the model prediction or to perform a first principles calculation to augment the training set of the model. The method is applied to a range of single- and multi-element systems and shown to achieve a favorable balance of accuracy and computational efficiency, while requiring a minimal amount of ab initio training data. We provide a fully open-source implementation of our method, as well as a procedure to map trained models to computationally efficient tabulated force fields.

36 MATERIALS SCIENCE↗
DOE OSTI2024

PHASM: A Toolkit for Creating AI Surrogate Models within Legacy Codebases

PHASM (“Parallel Hardware viA Surrogate Models”) is a software toolkit for creating AI-based surrogate models of scientific code. AI-based surrogate models are widely used for creating fast and inverse simulations. PHASM anticipates an additional future use case: adapting legacy code to modern hardware. While data centers are investing in heterogeneous hardware such as GPUs and FPGAs, many established scientific codebases remain unable to take advantage of the hardware’s higher parallelism without undergoing a costly rewrite. An alternative is to train a AI-based surrogate model to mimic computationally intensive functions in the code, and run the surrogate instead. PHASM formalizes a development lifecycle for such surrogate models, including discovering functions amenable to replacement with a surrogate model, predicting the resulting performance, identifying the function’s space of inputs and outputs, binding the model to the code, and managing model versions. A suite of software tools for facilitating these steps was written and validated against a set of model problems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Predicting Nugget Size of Resistance Spot Welds Using Infrared Thermal Videos With Image Segmentation and Convolutional Neural Network

Resistance spot welding (RSW) is a widely adopted joining technique in automotive industry. Recent advancement in sensing technology makes it possible to collect thermal videos of the weld nugget during RSW using an infrared (IR) camera. The effective and timely analysis of such thermal videos has the potential of enabling in situ nondestructive evaluation (NDE) of the weld nugget by predicting nugget thickness and diameter. Deep learning (DL) has demonstrated to be effective in analyzing imaging data in many applications. However, the thermal videos in RSW present unique data-level challenges that compromise the effectiveness of most pre-trained DL models. We propose a novel image segmentation method for handling the RSW thermal videos to improve the prediction performance of DL models in RSW. The proposed method transforms raw thermal videos into spatial-temporal instances in four steps: video-wise normalization, removal of uninformative images, watershed segmentation, and spatial-temporal instance construction. The extracted spatial-temporal instances serve as the input data for training a DL-based NDE model. The proposed method is able to extract high-quality data with spatial-temporal correlations in the thermal videos, while being robust to the impact of unknown surface emissivity. Overall, our case studies demonstrate that the proposed method achieves better prediction of nugget thickness and diameter than predicting without the transformation.

42 ENGINEERING↗
DOE OSTI2023

Optimizing Phase Change Composite Thermal Energy Storage Using the Thermal Ragone Framework

The thermal Ragone framework can be applied to the design of thermal energy storage heat exchangers to identifying relationships between their power requirements and energy capabilities. This presentation focuses on the design of planar thermal energy storage heat exchangers with phase change materials and thermal conductivity additives. Finite-difference models were used to optimize thermal storage heat exchanger designs focusing on key design parameters of conductivity additive volume fraction, spacing between tubes, and the phase transition temperatures. Design trends were developed for optimal designs based on varying thermal power requirements. To aid in future device design, models of reduced complexity were developed with the ability to predict optimal designs with computational speeds up to 5,000 times faster than finite-difference models.

ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION,↗
DOE OSTI2025

Catalytic resonance theory for parametric uncertainty of programmable catalysis

Microkinetic models are useful tools for screening catalytic materials; however, errors in their input parameters can lead to significant uncertainty in model predictions of catalyst performance. Here, in this work, we investigate the impact of linear scaling and Brønsted-Evans-Polanyi relation parametric uncertainty on microkinetic predictions of programmable-catalyst performance. Two case studies are considered: a generic A-to-B prototype reaction and the oxygen evolution reaction (OER). The results show that error-unaware models can accurately predict trends and, for the prototype reaction, values of optimal waveform parameters. The specific model parameters driving output uncertainty are identified via variance-based global sensitivity analysis. However, predictions of dynamic rate enhancement can decrease when uncertainty is propagated into the models. In both cases, we identify operating conditions where the programmable catalyst achieves a rate enhancement of at least one order of magnitude despite parametric uncertainty in the model, supporting programmable catalysis as a viable strategy for exceeding the Sabatier limit.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Editorial: Sorption processes in nuclear waste management: data knowledge management and new methodologies for data acquisition/prediction

A fundamental approach to Nuclear Waste Repository research involves the collection of experimental data in a laboratory setting, development of empirical and/or mechanistic numerical models representing those observations, and application of these models (or Reduced Order Models) into reactive transport and performance assessment models as predictive tools for informing society of impacts and risks associated with nuclear waste repository scenarios (Stevens et al., 2020). Therefore, the assimilation and interpretation of experimental data must take advantage of both new data and the rich historical data available in the literature and apply novel modeling approaches to improve predictive tools, particularly from the standpoint of uncertainty quantification, for nuclear waste repository performance assessment (Zavarin et al., 2022).

sorption↗
DOE OSTI2023

Comparing quantile regression forest and mixture density long short-term memory models for probabilistic post-processing of satellite precipitation-driven streamflow simulations

Abstract. Deep learning (DL) and machine learning (ML) are widely used in hydrological modelling, which plays a critical role in improving the accuracy of hydrological predictions. However, the trade-off between model performance and computational cost has always been a challenge for hydrologists when selecting a suitable model, particularly for probabilistic post-processing with large ensemble members. This study aims to systematically compare the quantile regression forest (QRF) model and countable mixtures of asymmetric Laplacians long short-term memory (CMAL-LSTM) model as hydrological probabilistic post-processors. Specifically, we evaluate their ability in dealing with biased streamflow simulations driven by three satellite precipitation products across 522 nested sub-basins of the Yalong River basin in China. Model performance is comprehensively assessed using a series of scoring metrics from both probabilistic and deterministic perspectives. Our results show that the QRF model and the CMAL-LSTM model are comparable in terms of probabilistic prediction, and their performances are closely related to the flow accumulation area (FAA) of the sub-basin. The QRF model outperforms the CMAL-LSTM model in most sub-basins with smaller FAA, while the CMAL-LSTM model has an undebatable advantage in sub-basins with FAA larger than 60 000 km2 in the Yalong River basin. In terms of deterministic predictions, the CMAL-LSTM model is preferred, especially when the raw streamflow is poorly simulated and used as input. However, setting aside the differences in model performance, the QRF model with 100-member quantiles demonstrates a noteworthy advantage by exhibiting a 50 % reduction in computation time compared to the CMAL-LSTM model with the same ensemble members in all experiments. As a result, this study provides insights into model selection in hydrological post-processing and the trade-offs between model performance and computational efficiency. The findings highlight the importance of considering the specific application scenario, such as the catchment size and the required accuracy level, when selecting a suitable model for hydrological post-processing.

Geology↗
DOE OSTI2023

An Online Prototype Toolset for Predicting and Optimizing P&T Performance (FY23 Status Report)

A new web-based toolset is being developed to support ongoing remediation optimization efforts and implementation of an adaptive site management strategy for the 200 West Area Pump-and-Treat (P&T) system at the Hanford Site. This toolset, comprising the well performance index tool and the well optimization pre-screening tool, will offer a user-friendly interface to predict and optimize the P&T well network’s performance at a preliminary level. Efforts in fiscal year (FY) 2023 focused on three main components: updating the existing deep learning model for predicting P&T performance, designing and developing a prototype of a web-based performance index tool, and initiating the conceptual design of the well optimization pre-screening tool. The well performance index tool is based on a pre-trained deep learning model that allows users to select a target contaminant and well screen length, then visualize the predicted performance of potential new wells across the site. The well optimization pre-screening tool includes two separate modules: the pre-computed scenario viewer, which organizes and visualizes offline optimization simulation results, and the quick analysis module, which provides real-time model prediction using user-specified well locations. In FY24, the plan is to add web-based applications to SOCRATES for both the well performance prediction tool and the optimization prescreening tool, with accompanying user and theory guides. These tools are intended to enable an accessible, easily applied, and transparent approach to remedy planning and decision-making.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Tula: Optimizing Time, Cost, and Generalization in Distributed Large-Batch Training

Distributed training increases the number of batches processed per iteration either by scaling-out (adding more nodes) or scaling-up (increasing the batch-size). However, the largest configuration does not necessarily yield the best performance. Horizontal scaling introduces additional communication overhead, while vertical scaling is constrained by computation cost and device memory limits. Thus, simply increasing the batch-size leads to diminishing returns: training time and cost decrease initially but eventually plateaus, creating a knee-point in the time/cost vs. batch-size pareto curve. The optimal batch-size therefore depends on the underlying model, data and available compute resources. Large batches also suffer from worse model quality due to the well-known “generalization gap”. In this paper, we present Tula, an online service that automatically optimizes time, cost, and convergence quality for large-batch training of convolutional models. It combines parallel-systems modeling with statistical performance prediction to identify the optimal batchsize. Tula predicts training time and cost within 7.5−14% error across multiple models, and achieves up to 20× overall speedup and improves test accuracy by ≈9% on average over standard large-batch training on various vision tasks, thus successfully mitigating the generalization gap and accelerating training at the same time.

Tyagi, Sahil [ORNL] (ORCID:0009000783144745)↗
DOE OSTI2021

Adaptive behavior and different thermal experiences of real people: A Bayesian neural network approach to thermal preference prediction and classification

Various observed and unquantifiable factors affect the thermal comfort of occupants in indoor environments and can lead to high uncertainty in the prediction and classification of their thermal preferences. The behavioral adaptation of occupants, by operating window systems for example, changes their thermal experience and expectations and therefore contributes to even higher prediction uncertainty. In this study, we applied a Bayesian neural network (BNN) algorithm to build a predictive model for occupant thermal preference using the ASHRAE Global Thermal Comfort Database II. The Bayesian method allows us to synthesize prior knowledge and available measurements into a unified modeling framework. It also offers a way to express and quantify uncertainty. Here we have performed a systematic study to test the efficiency and robustness of different BNN model configurations. In this study, the results show that the BNN model outperforms conventional thermal comfort models such as Predicted Mean Vote (PMV) and adaptive comfort model. The BNN model tends to produce more confident “prefer cooler” predictions with high possibility and low uncertainty. In contrast, the BNN model produces less certain predictions for “prefer no change” and “prefer warmer” across all occupants. Our findings suggest that linking occupants’ subjective evaluation measures and window opening/closing behavior to thermal comfort modeling effectively improves predictive performance.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

2023 Real-time Optimization Workflow Status Update

Nuclear integrated energy systems are composed of a diverse set of energy generation sources and exist in dynamic and competitive electricity markets. With the inclusion of thermal energy storage, nuclear power systems can store heat for future use through various processes, such as water desalination or hydrogen generation. This heat storage can be managed in such a way as to economically optimize its usage. By combining real-time price data from the day-ahead and real-time markets with predictive and intelligent models, the charging and discharging of the thermal energy storage may be determined and optimized. This research details an approach through models and systems for the real-time optimization (RTO) of capacity allocation. Virtual models of the energy system and its physics phenomenon and component interactions provide intelligence to verification and prediction of operations. An optimization framework can use data generated from both a set of physical assets as well as the virtual models to predict future performance and create optimization and control workflows. Data warehouse technologies can be used to combine data across models, optimization workflows, market price data, and sensor data to intuitively store various types of data and provide integrations to physical control systems as well as user visualizations. Put together, these components can create a system for the RTO of nuclear integrated energy systems. Various virtual models and bench-scale physical demonstrations have been successfully performed and verified using this system. Larger scale testbeds that include thermal energy storage systems have been identified as future opportunities. A gap analysis that details the steps necessary to reach a physical demonstration at this scale is provided, along with conclusions on the current effort and future work.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Utilizing integrated neutron diffraction and elastoplastic self-consistent crystal plasticity model to quantitatively assess the strengthening mechanism in Al–12.5Ce and Al–12.5Ce–0.4Mg alloys

An integrated in-situ neutron diffraction and elastic plastic self-consistent crystal plasticity (EPSC-CP) modeling scheme is performed on a binary Al–12Ce alloy and a ternary Al–12Ce–0.4Mg alloys. Using this scheme, the constitutive parameters, i.e. elastic constants and slip system parameters of individual phases can be calibrated which can be used in microstructure-based CP models to predict materials performance. From this study, it is shown that the elastic constants of Al 11 Ce 3 intermetallics calculated from density function theory calculation in the literature are rather accurate. When applied to the EPSC-CP model, the lattice strains of both the binary and ternary alloys are correctly predicted as compared with experiments, and large lattice strain differences between Al (100) plane and Al 11 Ce 3 (010) plane are demonstrated. The slip system parameters calibrated by the scheme shows that the addition of 0.4 wt% Mg in the alloy has little influence on the critical resolved shear stress of initial dislocation glide in the Al matrix which caused plastic yield in the material. This can be explained by the very dilute Mg solute content in the Al solid solution, causing large spacing of Al–Mg lattice misfit sites and little impact on resistance of dislocation glide at initial yield. The 0.4 wt% Mg addition, on the other hand, has a large influence on the hardening term in the slip system parameters, indicating those Al–Mg misfit sites do help dislocation accumulation during the deformation. The impact of dilute Mg addition on the Al slip system parameters is also reflected in the flow behavior of the ternary alloy: little impact on the yield stress, but a large impact on working hardening and tensile strength of the materials which is consistent with the literature.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Machine learning elastic constants of multi-component alloys

The present manuscript explores application of machine learning methods for determining elastic constants and other derived mechanical properties of multi-component alloys. Here, a number of machine learning models, including linear regression, neural network and random forest based models, are trained and tested on a dataset of binary alloys generated using density functional theory (DFT) calculations and spanning over a large number of elemental species in the periodic table. Starting with a wide range of simple and easily accessible compositionally-averaged elemental features, a correlation-based feature selection strategy was used to systematically down-select a set of most relevant features towards the prediction of the elasticity tensor components. The true predictive performance and the associated uncertainties of the models were established by testing on unseen data and bootstrapping, respectively. A single and pair-wise feature partial dependence analysis was performed to visualize the average property trends in the multi-dimensional feature space in order to further understand the achieved predictive performance. The utility of the trained model is further demonstrated by obtaining sufficiently accurate yet highly efficient approximations for bulk modulus, Young’s modulus, shear modulus and Poisson’s ratio for alloys beyond the binary space (i.e., two-component alloys) on which the model was originally trained. More importantly, we test and validate the predictive performance of the developed model directly against the experimentally measured elastic constants of technologically relevant multi-component alloys (such as, Ni- and Ti-based alloys). Finally, utility of such a data-enabled route is demonstrated by predicting the possible range of various elastic properties for vast composition space available within the five component Ni-Cr-Fe-Mo-W alloy system in a high-throughput manner.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Application of machine learning to evaluating and remediating models for energy and environmental engineering

Machine learning (ML) algorithms have been increasingly successful in their applications to solve energy and environmental engineering problems. ML algorithms have the advantage of being able to solve highly nonlinear issues effectively. Furthermore, considering the limited sample size of data collected in energy and environmental engineering, obtaining a ML model with reasonable accuracy is simple. Unfortunately, the vast majority of the current applications of ML algorithms lack effective screening of dominant factors and comprehensive model validation, which weakens the predictive ability of the models. The present study takes the minimum miscible pressure (MMP) of CO 2 - oil systems as an example. It establishes a systematic and robust predictive model to address this issue. Based on 147 sets of slim tube tests, the predictive models of the MMPs are investigated by application of eight ML algorithms. The paper concludes that most of the published ML models in the field of energy and environmental engineering prediction are not reliable. Furthermore, it addresses the main reasons for the poor performance of some predictive models built by ML and provides guidelines on how to make such models robust. Further, to the best of our knowledge, this is the first study to point out the defects of current ML modeling methods and propose countermeasures for their application in energy and environmental engineering problems.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2023

Enabling site-specific well leakage risk estimation during geologic carbon sequestration using a modular deep-learning-based wellbore leakage model

Geologic carbon sequestration (GCS) is a promising technology for mitigating net carbon emissions and growing climate concern by storing CO 2 in reservoirs. Oil and gas brownfields are an attractive option for CO 2 storage, but these sites have many historical wellbores from petroleum production and can be a potential leakage pathway for CO 2 or formation brine. Therefore, risk management of GCS operations requires an assessment of potential well leakage. Due to the high uncertainty of the system, stochastic approaches are ideal for quantifying the range of risk behaviors, but they must be computationally efficient in the face of complex physics. Here, we develop a new physics-centric deep learning wellbore model to predict the leakage of CO 2 and brine through leaky wellbores. Multi-physics numerical simulations were used to generate data sets, and physics-informed features were introduced. Neural networks were optimized with an automated searching algorithm. Feature analysis quantifies the impact of each feature on model prediction and confirms the role of physics-inspired parameters. The model shows high predictive performance across a wide range of geologic and injection conditions and well attributes. In conclusion, a case study illustrates how the model is applied to assess well leakage in GCS operations.

58 GEOSCIENCES↗
DOE OSTI2021

Modeling Coupled Chemo-Mechanical Behavior of Randomly Oriented NMC811 Polycrystalline Li-Ion Battery Cathodes

This paper develops a three-dimensional, transient, chemo-mechanical model that predicts the performance of single secondary particle Li-ion battery cathodes. The secondary particles are composed of numerous (approximately 60) randomly oriented single-crystal primary particles. The model incorporates concentration-dependent and anisotropic material properties. As much as possible, electrochemical, transport, and structural properties for crystalline NMC811 (Li x Ni 0.8 Mn 0.1 Co 0.1 O 2 ) are taken from prior publications. Weak Van der Waals bonding between primary particles is modeled empirically using a spring analogy, which enables local primary-particle separations (disintegration) and subsequent reattachments. The model fully couples Li diffusion and the mechanical response. Results include predictions of local Li-concentrations and stresses. High stresses are found near grain boundaries, especially when the lattice orientations are greatly misaligned. Particle separations are characterized in terms of a damage parameter. The model is used to predict the effects of design and operating conditions, including charge/discharge rates, cycling scenarios, and particle sizes.

25 ENERGY STORAGE↗