Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “performance 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 1,135 records · Page 63
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Field Performance of Commercial Building Load Flexibility Using Model Predictive Control

Model Predictive Control (MPC) applied to buildings is starting to see some commercial adoption by companies. However, it is hard to estimate if relative energy cost savings are enough to justify the cost of MPC implementation with few reported demonstrations. In small commercial and residential buildings, a one size-fits-all solution can help reduce implementation costs, while in very large buildings or districts the potential energy cost savings magnitude can cover more tailored solutions. This estimation becomes harder for medium to large commercial buildings, where a one-size-fits-all solution cannot be adopted and potential energy cost savings might not be sufficient to cover a tailored solution. Therefore, value propositions in addition to energy efficiency alone can make MPC technology more attractive through additional energy cost savings. One such value proposition is load shifting in response to dynamic electricity prices. On this aspect, MPC is a key technology to unlock building thermal mass for energy flexibility in response to electric grid conditions. This study shows the experimental results of MPC control of an office building in Berkeley, where different dynamic electricity price profiles were used in the MPC objective function to shift the building load and to calculate hypothetical electricity costs. Results show potential 50% cost savings with respect to the existing controller with the dynamic price scenario.

Zanetti, Ettore↗
DOE OSTI2024

Implementing a Laser Stabilization System for Trapping Ca+ Ions: an Internship Reflection

At Lawrence Livermore National Laboratory, I contributed to a project developing 3D printed micro ion traps for quantum computing. I designed, implemented, and assessed a laser stabilization system that locked lasers to the frequencies required for calibrating our High Finesse WS8-10 wavelength meter and for laser cooling and trapping of Ca+ ions. I also programmed a Python interface for hardware communication, data collection, and statistical analysis. Additionally, I optimized and aligned laser beam paths, and I implemented a closed digital feedback loop using Proportional, Integral, and Derivative (PID) control parameters. I analyzed both the long-term and short-term behavior of our locked lasers and adjusted PID parameters to enhance performance. Furthermore, I used COMSOL to simulate the capacitance of a linear Paul trap design and predict our trap’s performance. The procedures I developed for the interface, analysis, and simulations will continue to support the ion trapping experiment after my appointment. I strengthened my skills in data analysis, Python coding, and optical alignment for laser systems. My confidence as a researcher grew, particularly in communicating my research. This experience taught me the importance of careful planning and consideration in research and solidified my desire to continue exploring novel quantum technology as an undergraduate

42 ENGINEERING↗
DOE OSTI2024

Tandem Predictions for HPC Jobs

At the core of the predictive analytics applied to High Performance Computing (HPC), the most prominent tasks are the prediction of job runtimes and the prediction of job queue times, both of which have the potential for informing HPC users during their every-day decision making. Accurate runtime predictions can help users better choose so-called wallclock times at job submission, decreasing the odds of their jobs waiting in queues longer than necessary. The accurate and timely queue time predictions offered for the available partitions can inform the favorable selection of partitions for running jobs. This potential is well understood as we see in the abundance of research studies that propose solutions for these tasks, including the work published in the last several years. These tasks are seemingly receptive to the Machine Learning (ML) solutions, considering that there is no shortage of training data where HPC centers over time run millions and millions of jobs. However, we study the existing research literature, as well as look for examples in the toolchains supported on the exemplar HPC facilities, and, surprisingly, do not find any practical solutions that are ready to be adopted. We interpret this as a manifestation of the shortage of UX/UI efforts that support HPC analytics and also as a sign that the research has not come to the consensus on solving these tasks. In this study, we aim to shed new light on the long-running task of job queue time prediction by exploring the utility of runtime predictions in improving prediction accuracy and, actually, predicting these two metrics together, in tandem. In other words, we show how runtime predictions become valuable input in the queue time modeling. We challenge the existing approaches to feature engineering for the queue time prediction and describe promising results we obtained for a large dataset of HPC jobs from a supercomputer at the National Renewable Energy Laboratory.

HPC↗
DOE OSTI2025

Tandem Predictions for HPC Jobs: Preprint

At the core of the predictive analytics applied to High Performance Computing (HPC), the most prominent tasks are the prediction of job runtimes and the prediction of job queue times, both of which have the potential for informing HPC users during their every-day decision making. Accurate runtime predictions can help users better choose so-called wallclock times at job submission, decreasing the odds of their jobs waiting in queues longer than necessary. The accurate and timely queue time predictions offered for the available partitions can inform the favorable selection of partitions for running jobs. This potential is well understood as we see in the abundance of research studies that propose solutions for these tasks, including the work published in the last several years. These tasks are seemingly receptive to the Machine Learning (ML) solutions, considering that there is no shortage of training data where HPC centers over time run millions and millions of jobs. However, we study the existing research literature, as well as look for examples in the toolchains supported on the exemplar HPC facilities, and, surprisingly, do not find any practical solutions that are ready to be adopted. We interpret this as a manifestation of the shortage of UX/UI efforts that support HPC analytics and also as a sign that the research has not come to the consensus on solving these tasks. In this study, we aim to shed new light on the long-running task of job queue time prediction by exploring the utility of runtime predictions in improving prediction accuracy and, actually, predicting these two metrics together, in tandem. In other words, we show how runtime predictions become valuable input in the queue time modeling. We challenge the existing approaches to feature engineering for the queue time prediction and describe promising results we obtained for a large dataset of HPC jobs from a supercomputer at the National Renewable Energy Laboratory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Investigating the relationship between radial pre-stress magnitude and ballistic projectile dwell in heavy confined ceramic targets

A series of bare silicon carbide ceramic armour disks were manufactured with high-strength steel containment to induce varying magnitudes of radial pre-stress. The induced radial stresses, ranging from 0 to 900 MPa, were measured via neutron diffraction and verified by comparison with numerical calculations. Four targets of each configuration, referred to as slip-fit, moderate pre-stress, and high pre-stress, were subject to ballistic testing with hemispherical-nose, tungsten heavy alloy long rod projectiles to determine the interface defeat transition velocity. In-situ diagnostics were unsuccessful in aiding the identification of interface defeat, necessitating a reliance on post-mortem assessment. A transition velocity of approx. 1000 m/s was identified for the unstressed target, increasing to approx. 1200 m/s for the pre-stressed configurations. No performance effect was discernible between the moderate (372 MPa) and high (899 MPa) pre-stress configurations, suggesting that an optimal performance may be achieved for lower pre-stress levels (i.e., <372 MPa). Additionally, the test results were compared with four semi-analytical predictions of interface defeat performance and good agreement was found, albeit with significant range in the model predictions.

42 ENGINEERING↗
DOE OSTI2020

Multi-Scale Simulation of Wind Farm Performance during a Frontal Passage

Predicting the response of wind farms to changing flow conditions is necessary for optimal design and operation. Here in this work, simulation and analysis of a frontal passage through a utility scale wind farm is achieved for the first time using a seamless multi-scale modeling approach. A generalized actuator disk (GAD) wind turbine model is used to represent turbine–flow interaction, and results are compared to novel radar observations during the frontal passage. The Weather Research and Forecasting (WRF) model is employed with a nested grid setup that allows for coupling between multi-scale atmospheric conditions and turbine response. Starting with mesoscale forcing, the atmosphere is dynamically downscaled to the region of interest, where the interaction between turbulent flows and individual wind turbines is simulated with 10 m grid spacing. Several improvements are made to the GAD model to mimic realistic turbine operation, including a yawing capability and a power output calculation. Ultimately, the model is able to capture both the dynamics of the frontal passage and the turbine response; predictions show good agreement with observed background velocity, turbine wake structure, and power output after accounting for a phase shift in the mesoscale forcing. This study demonstrates the utility of the WRF-GAD model framework for simulating wind farm performance under complex atmospheric conditions.

17 WIND ENERGY↗
DOE OSTI2023

Deep learning integrates histopathology and proteogenomics at a pan-cancer level

We introduce a pioneering approach that integrates pathology imaging with transcriptomics and proteomics to identify predictive histology features associated with critical clinical outcomes in cancer. We utilize 2,755 H&E-stained histopathological slides from 657 patients across 6 cancer types from CPTAC. Our models effectively recapitulate distinctions readily made by human pathologists: tumor vs. normal (AUROC = 0.995) and tissue-of-origin (AUROC = 0.979). We further investigate predictive power on tasks not normally performed from H&E alone, including TP53 prediction and pathologic stage. Importantly, we describe predictive morphologies not previously utilized in a clinical setting. The incorporation of transcriptomics and proteomics identifies pathway-level signatures and cellular processes driving predictive histology features. Model generalizability and interpretability is confirmed using TCGA. We propose a classification system for these tasks, and suggest potential clinical applications for this integrated human and machine learning approach. A publicly available web-based platform implements these models.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

Laser-system model for enhanced operational performance and flexibility on OMEGA EP

The development of laser performance models having real-time prediction capability for the OMEGA EP laser system has been essential in meeting requests from its user community for increasingly complex pulse shapes that span a wide range of energies. The laser operations model PSOPS provides rapid and accurate predictions of OMEGA EP laser-system performance in both forward and backward directions, a user-friendly interface and rapid optimization capability between shots. We describe the model’s features and show how PSOPS has allowed real-time optimization of the laser-system configuration in order to satisfy the demands of rapidly evolving experimental campaign needs. We also discuss several enhancements to laser-system performance accuracy and flexibility enabled by PSOPS.

42 ENGINEERING↗
DOE OSTI2020

Beyond Idealized Models of Nanoscale Metal Hydrides for Hydrogen Storage

Metal hydrides are attractive for compact, low-pressure hydrogen storage, yet a foundational understanding of factors governing their thermodynamics and kinetics is still lacking. Predictive modeling from the atomic to the microstructural scale plays a critical role in addressing these gaps, particularly for nanoscale materials, which promise improved performance but are difficult to probe. In this paper, we summarize strategies being developed within the Hydrogen Materials—Advanced Research Consortium (HyMARC) for going beyond conventional models to incorporate more complex physics, more realistic structures, and better approximation of operation conditions in simulations of nanoscale metal hydrides. We highlight four beyond-ideal factors that influence predicted performance: (1) surface anharmonic dynamics, (2) interface and surface energy penalties, (3) mechanical stress under confinement, and (4) the presence of native surface oxide. Approaches for addressing these factors are demonstrated on model materials representative of high-capacity hydrogen storage systems, and implications for understanding performance under operating conditions are discussed.

08 HYDROGEN↗
DOE OSTI2022

A systematic method for selecting molecular descriptors as features when training models for predicting physiochemical properties

Machine learning has proven to be a powerful tool for accelerating biofuel development. Although numerous models are available to predict a range of properties using chemical descriptors, there is a trade-off between interpretability and performance. Neural networks provide predictive models with high accuracy at the expense of some interpretability, while simpler models such as linear regression often lack in accuracy. In addition to model architecture, feature selection is also critical for developing interpretable and accurate predictive models. We present a method for systematically selecting molecular descriptor features and developing interpretable machine learning models without sacrificing accuracy. Our method simplifies the process of selecting features by reducing feature multicollinearity and enables discoveries of new relationships between global properties and molecular descriptors. To demonstrate our approach, we developed models for predicting melting point, boiling point, flash point, yield sooting index, and net heat of combustion with the help of the Tree-based Pipeline Optimization Tool (TPOT). For training, we used publicly available experimental data for up to 8351 molecules. Our models accurately predict various molecular properties for organic molecules (mean absolute percent error (MAPE) ranges from 3.3% to 10.5%) and provide a set of features that are well-correlated to the property. This method enables researchers to explore sets of features that significantly contribute to the prediction of the property, offering new scientific insights. To help accelerate early stage biofuel research and development, we also integrated the data and models into a open-source, interactive web tool.

09 BIOMASS FUELS↗
DOE OSTI2024

AstroCLIP: a cross-modal foundation model for galaxies

ABSTRACT We present AstroCLIP, a single, versatile model that can embed both galaxy images and spectra into a shared, physically meaningful latent space. These embeddings can then be used – without any model fine-tuning – for a variety of downstream tasks including (1) accurate in-modality and cross-modality semantic similarity search, (2) photometric redshift estimation, (3) galaxy property estimation from both images and spectra, and (4) morphology classification. Our approach to implementing AstroCLIP consists of two parts. First, we embed galaxy images and spectra separately by pre-training separate transformer-based image and spectrum encoders in self-supervised settings. We then align the encoders using a contrastive loss. We apply our method to spectra from the Dark Energy Spectroscopic Instrument and images from its corresponding Legacy Imaging Survey. Overall, we find remarkable performance on all downstream tasks, even relative to supervised baselines. For example, for a task like photometric redshift prediction, we find similar performance to a specifically trained ResNet18, and for additional tasks like physical property estimation (stellar mass, age, metallicity, and specific-star-formation rate), we beat this supervised baseline by 19 per cent in terms of R2. We also compare our results with a state-of-the-art self-supervised single-modal model for galaxy images, and find that our approach outperforms this benchmark by roughly a factor of two on photometric redshift estimation and physical property prediction in terms of R2, while remaining roughly in-line in terms of morphology classification. Ultimately, our approach represents the first cross-modal self-supervised model for galaxies, and the first self-supervised transformer-based architectures for galaxy images and spectra.

Parker, Liam (ORCID:0009000749521674)↗
DOE OSTI2023

Performance analysis and comparison of data-driven models for predicting indoor temperature in multi-zone commercial buildings

Building thermal models, which characterize the properties of a building’s envelope and thermal mass, are essential for accurate indoor temperature and cooling/heating demand prediction. Because of their flexibility and ease of use, data-driven models are increasingly used. Here, this study compared and analyzed the performance of gray-box (resistance-capacitance) and black-box (recurrent neural network) models for predicting indoor air temperature in a real multi-zone commercial building. The developed resistance-capacitance model served as a benchmark model for which full sets of temporal data and building information were used as inputs. The recurrent neural network models were trained and tested assuming various available types and amounts of temporal data and known building physical information to investigate the effects of data and information availability. Feature importance analysis was conducted to select the key variables for different prediction targets under different scenarios. This research provides guidance in selecting an appropriate building thermal response modeling method based on the measured data availability, building physical information, and application.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2020

Computational discovery of promising new n-type dopable ABX Zintl thermoelectric materials

Computational prediction of good thermoelectric (TE) performance in several n-type doped Zintl phases, combined with successful experimental realization, has sparked interest in discovering new n-type dopable members of this family of materials. However, most known Zintls are typically only p-type dopable; prior successes in finding n-type Zintl phases have been largely serendipitous. Here in this paper, we go beyond previously synthesized Zintl phases and perform chemical substitutions in known n-type dopable ABX Zintl phases to discover new ones. We use first-principles calculations to predict their stability, potential for TE performance as well as their n-type dopability. Using this approach, we find 17 new ABX Zintl phases in the KSnSb structure type that are predicted to be stable. Several of these newly predicted phases (KSnBi, RbSnBi, NaGeP) are found to exhibit promising n-type TE performance and are n-type dopable. We propose these compounds for further experimental studies, especially KSnBi and RbSnBi, which are both predicted to be good TE materials with high electron concentrations due to self-doping by native defects, when grown under alkali-rich conditions.

36 MATERIALS SCIENCE↗
DOE OSTI2021

DeepDist: real-value inter-residue distance prediction with deep residual convolutional network

Abstract Background Driven by deep learning, inter-residue contact/distance prediction has been significantly improved and substantially enhanced ab initio protein structure prediction. Currently, most of the distance prediction methods classify inter-residue distances into multiple distance intervals instead of directly predicting real-value distances. The output of the former has to be converted into real-value distances to be used in tertiary structure prediction. Results To explore the potentials of predicting real-value inter-residue distances, we develop a multi-task deep learning distance predictor (DeepDist) based on new residual convolutional network architectures to simultaneously predict real-value inter-residue distances and classify them into multiple distance intervals. Tested on 43 CASP13 hard domains, DeepDist achieves comparable performance in real-value distance prediction and multi-class distance prediction. The average mean square error (MSE) of DeepDist’s real-value distance prediction is 0.896 Å 2 when filtering out the predicted distance ≥ 16 Å, which is lower than 1.003 Å 2 of DeepDist’s multi-class distance prediction. When distance predictions are converted into contact predictions at 8 Å threshold (the standard threshold in the field), the precision of top L/5 and L/2 contact predictions of DeepDist’s multi-class distance prediction is 79.3% and 66.1%, respectively, higher than 78.6% and 64.5% of its real-value distance prediction and the best results in the CASP13 experiment. Conclusions DeepDist can predict inter-residue distances well and improve binary contact prediction over the existing state-of-the-art methods. Moreover, the predicted real-value distances can be directly used to reconstruct protein tertiary structures better than multi-class distance predictions due to the lower MSE. Finally, we demonstrate that predicting the real-value distance map and multi-class distance map at the same time performs better than predicting real-value distances alone.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Utilization of Carbon Supply Chain Wastes and Byproducts to Manufacture Graphite for Energy Storage Applications

This project explored how coal and waste coal can be transformed into high-value graphite used in batteries for electric vehicles and power grids. A manufacturing technique that converts coal into carbon foam and subsequently graphite was developed and refined to improve the thermal and mechanical properties of the resulting materials. Cost-effectiveness and environmental impacts were also evaluated. Coal-derived graphite, especially when processed at high temperatures, was shown to perform comparably to commercial graphite in battery tests. Advanced computer simulations were used to model battery behavior and predict long-term performance, demonstrating that coal-based graphite has the potential to support electric vehicle deployment while reducing reliance on imported materials. This work offers a promising pathway for cleaner energy storage solutions and creates economic opportunities for coal-reliant communities by generating new uses for mining byproducts.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2023

Efficient Optimization of Energy Recovery From Geothermal Reservoirs With Recurrent Neural Network Predictive Models

Improving the long-term energy production performance of geothermal reservoirs can be accomplished by optimizing field development and management plans. Reliable prediction models, however, are needed to evaluate and optimize the performance of the underlying reservoirs under various operation and development strategies. In traditional frameworks, physics-based simulation models are used to predict the energy production performance of geothermal reservoirs. However, detailed simulation models are not trivial to construct, require a reliable description of the reservoir conditions and properties, and entail high computational complexity. Data-driven predictive models can offer an efficient alternative for use in optimization workflows. This paper presents an optimization framework for net power generation in geothermal reservoirs using a variant of the recurrent neural network (RNN) as a data-driven predictive model. The RNN architecture is developed and trained to replace the simulation model for computationally efficient prediction of the objective function and its gradients with respect to the well control variables. The net power generation performance of the field is optimized by automatically adjusting the mass flow rate of production and injection wells over 12 years, using a gradient-based local search algorithm. Two field-scale examples are presented to investigate the performance of the developed data-driven prediction and optimization framework. Furthermore, the prediction and optimization results from the RNN model are evaluated through comparison with the results obtained by using a numerical simulation model of a real geothermal reservoir.

15 GEOTHERMAL ENERGY↗
DOE OSTI2021

Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications

Memory design space exploration methods study memory systems’ performances and limitations before implementation. The computer memory design space has grown exponentially because of the enormous growth of memory types, memory controllers, and application software. Computer simulators are commonly used for memory design space exploration. However, complex memory simulations take an enormous amount of time. Hence, in this paper, we proposed a machine learning-based design space exploration method for dynamic random-access memory and non-volatile memory systems. We applied our method to the CosmoGAN and LeNet applications to predict the following six memory response parameters: (i) bandwidth, (ii) power, (iii) average latency, (iv) average total latency, (v) memory reads, and (vi) memory writes. Our experimental results show that machine learning models can predict memory response parameter values faster than simulations. We used support vector machine, random forest, and gradient boosting machine learning models. We observed that the support vector machine provides better performance for bandwidth, average latency, and average total latency. The random forest model works better for memory reads and writes. The gradient boosting model provides superior prediction performance for power. We provide a detailed discussion on learning curve characteristics, error analysis, and memory type recommendation.

Hasan, S M Shamimul↗
DOE OSTI2022

Proteome-scale Deployment of Protein Structure Prediction Workflows on the Summit Supercomputer

Deep learning has contributed to major advances in the prediction of protein structure from sequence, a fundamental problem in structural bioinformatics. With predictions now approaching the accuracy of crystallographic experiments, and with accelerators like GPUs and TPUs making inference using large models rapid, genome-level structure prediction becomes an obvious aim. Leadership-class computing resources can be used to perform genome-scale protein structure prediction using state-of-the-art deep learning models, providing a wealth of new data for systems biology applications. Here we describe our efforts to efficiently deploy the AlphaFold v.2 program, for full-proteome structure prediction, at scale on the Oak Ridge Leadership Computing Facility's resources, including the Summit supercomputer. We performed inference to produce the predicted structures for 40,526 protein sequences, corresponding to four prokaryotic proteomes and one plant proteome, using under 4,400 total Summit node hours, equivalent to using the majority of the supercomputer for a little over one hour. We also designed an optimized structure refinement that reduced the time for the relaxation stage of the AlphaFold pipeline by over 10X for longer sequences. We demonstrate the types of analyses that can be performed on proteome-scale collections of sequences, including a search for novel quaternary structures and implications for functional annotation.

Gao, Mu↗