Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “real time optimization”

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 1,081 records · Page 60
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Sequential Decision Making (SDM) for Mesh Refinement and Model Selection in Multiscale, Multi-Physics Applications

Intelligent automation and decision support are needed to enhance computational efficiency and robustness in multiscale and multi-physics problems, including materials science, manufacturing, and climate and weather modeling. Current scientific computing approaches for enabling decisions by scientists fail to explore the role of learning, reasoning, and probabilistic planning. Often these decisions are not performed in real-time during the computation but are made prior to the start of the computation, which must be interrupted in order to make changes to the prior choices. Such interruptions at different stages of the computation increase the total computing time and the need for a human expert to frequently monitor the results. State of art scientific computing methods consist of rule-based algorithms that cannot automatically adapt to a dynamically changing computing environment. The development of a Sequential Decision Making (SDM) framework will automate scientific computing by optimizing the policies for mesh refinement, time-stepping, model and algorithm selection, resource allocation, and pre and post-processing. Our agent SDM framework for scientific computing will consist of data-driven learning (Classifier), automated reasoning (contextual knowledge), and probabilistic planning (Reinforcement Learning). In this project, we focused on three problems to demonstrate our SDM framework on a set of ordinary and partial differential equations. Classification of Lorenz system regions using Feed-Forward Neural Networks examined learning in the SDM framework. On the other hand, reasoning and planning in the SDM framework were used in two problems: adaptive time-stepping for nonlinear ODEs using on-policy RL algorithms, and adaptive mesh refinement for 2-D PDEs using off-policy RL algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Development of a Discrepancy Checker for the Digital Twin in a Supervisory Control System for a Thermal Energy Delivery System

Defined as a virtual representation of a physical object, process, or service, and used to support real-world decision-making, a digital twin (DT) can be utilized to combine classical and novel frameworks in sensors, state predictions, and multi-input/multi-output systems, and to enable optimal autonomous operations. However, a DT’s usefulness largely depends on its ability to adequately mirror the state of its physical counterpart, and this adequacy should be reflected by the level of uncertainty in the underlying simulation models when estimating and predicting quantities of interest (QOIs). Moreover, simulation models in a DT may involve multiple fidelities of representations—ranging from physics-based models to data-driven ones—but classical uncertainty quantification (UQ) methods struggle to handle numerous uncertainty sources, nor are they designed for real-time applications. This work presents a UQ-based discrepancy checking and diagnosis tool for a DT-based supervisory control system applied to a thermal energy delivery system (TEDS) at Idaho National Laboratory. The discrepancy checker was developed using metadata from an automated DT development process, and these metadata included different combinations of physical model forms and model parameters, training data and hyperparameters for surrogate models, and design parameters for supervisory control systems. Next, correlations between the uncertainty results and the metadata were established and then applied to the DT operations. The discrepancy checker evaluates the discrepancies between model predictions from virtual and sensor measurements and backtraces them to the corresponding major sources of uncertainty. The discrepancy checker showed reasonable performance in detecting discrepancies and diagnosing sources of uncertainty in testing scenarios.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2022

Atikokan Digital Twin: Machine learning in a biomass energy system

The Atikokan Generating Station, operated by Ontario Power Generation, has a 200 MW, biomass-fired tower boiler that operates on a dispatch schedule with a five-minute cycle. The boiler is generally operated in the range of 40–100 MW using two of five burner levels. In order to optimize boiler performance, we propose the implementation of a unique digital twin. Our digital twin abstraction couples Bayesian inference from science-based models and from observations (machine learning) with decision theory to predict operating-variable set points that optimize the physical asset (the boiler) in the presence of uncertainty (artificial intelligence). We focus this paper on the continuous Bayesian machine learning part of the Atikokan Digital Twin; we discuss decision theory in a companion paper. We identify and learn about 12 operational, model, and measured-output parameters and their uncertainties from high-fidelity, science-based simulations of the Atikokan boiler and from the observed measurements at the power plant. Since the goal of the Atikokan Digital Twin is to implement it online in real time, we require fast function evaluations for the quantities of interest extracted from the simulations in the Bayesian analysis. We use Gaussian process regression/interpolation to create accurate, robust surrogate models. We define the Bayesian priors and likelihood function and solve for the posterior distributions of the 12 parameters. Here we then propagate these distributions (i.e., parameters with uncertainty) into the predicted distributions of 790 quantities of interest to learn about the relative importance of various sources of error including experimental, model, and operating-parameter errors.

09 BIOMASS FUELS↗
DOE OSTI2020

High-pressure ignition delay time measurements of a four-component gasoline surrogate and its high-level blends with ethanol and methyl acetate

Ignition delay times of a gasoline surrogate (iso-octane/toluene/n-heptane/1-hexene at 55/25/15/5% by liquid volume, developed in McCormick et al. 2017) with high-level bioblendstock-gasoline surrogate blends (50% and 85% biofuel by liquid volume of each ethanol and methyl acetate) were collected behind reflected shock waves. Post-reflected-shock temperatures ranged from 968 to 1361 K at pressures of about 4, 10, 25, and 50 atm. Data were collected for real fuel–air mixtures at fuel lean and stoichiometric conditions (φ = 0.5, 1.0) with focus on the more-practical, fuel-lean conditions. Ignition delay times were measured from OH* chemiluminescence around 307 nm using an endwall diagnostic. The RON and MON of the surrogate were 90.3 and 84.7, respectively. Ethanol was chosen due to its wide use in flex-fuel vehicles and availability, while also providing an increased octane rating. Methyl acetate was chosen for its especially high octane rating to investigate the effect of an extreme case. To validate the gasoline surrogate, the data are compared to real gasoline (RD387) and several gasoline surrogate experiments from the literature. Using the wide range of pressures studied, the gasoline surrogate’s pressure dependence was quantified to account for test-to-test variations using regression analysis. Similarly, a global correlation for gasoline and its surrogates was developed using all available data from the literature. Two modern chemical kinetics models targeting gasoline and its surrogates are compared to the ignition delay time measurements. These new high-pressure, high-bioblendstock concentration tests provide required chemical kinetic data for optimizing fuel and engine design.

42 ENGINEERING↗
DOE OSTI2022

Universal control of a six-qubit quantum processor in silicon

Future quantum computers capable of solving relevant problems will require a large number of qubits that can be operated reliably. However, the requirements of having a large qubit count and operating with high fidelity are typically conflicting. Spins in semiconductor quantum dots show long-term promise but demonstrations so far use between one and four qubits and typically optimize the fidelity of either single- or two-qubit operations, or initialization and readout. Here, we increase the number of qubits and simultaneously achieve respectable fidelities for universal operation, state preparation and measurement. We design, fabricate and operate a six-qubit processor with a focus on careful Hamiltonian engineering, on a high level of abstraction to program the quantum circuits, and on efficient background calibration, all of which are essential to achieve high fidelities on this extended system. State preparation combines initialization by measurement and real-time feedback with quantum-non-demolition measurements. These advances will enable testing of increasingly meaningful quantum protocols and constitute a major stepping stone towards large-scale quantum computers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

High-throughput feedback-enabled optogenetic stimulation and spectroscopy in microwell plates

Abstract The ability to perform sophisticated, high-throughput optogenetic experiments has been greatly enhanced by recent open-source illumination devices that allow independent programming of light patterns in single wells of microwell plates. However, there is currently a lack of instrumentation to monitor such experiments in real time, necessitating repeated transfers of the samples to stand-alone analytical instruments, thus limiting the types of experiments that could be performed. Here we address this gap with the development of the optoPlateReader (oPR), an open-source, solid-state, compact device that allows automated optogenetic stimulation and spectroscopy in each well of a 96-well plate. The oPR integrates an optoPlate illumination module with a module called the optoReader, an array of 96 photodiodes and LEDs that allows 96 parallel light measurements. The oPR was optimized for stimulation with blue light and for measurements of optical density and fluorescence. After calibration of all device components, we used the oPR to measure growth and to induce and measure fluorescent protein expression in E. coli . We further demonstrated how the optical read/write capabilities of the oPR permit computer-in-the-loop feedback control, where the current state of the sample can be used to adjust the optical stimulation parameters of the sample according to pre-defined feedback algorithms. The oPR will thus help realize an untapped potential for optogenetic experiments by enabling automated reading, writing, and feedback in microwell plates through open-source hardware that is accessible, customizable, and inexpensive.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Optimizing Optical Searches for Supermassive Black Hole Binaries in Active Galactic Nuclei Light Curves: Fourier versus Bayesian Periodicity Detection

Simulations predict that supermassive black hole binaries (SMBHBs) will exhibit periodic brightness variations that may exceed the stochastic variability intrinsic to active galactic nuclei (AGN). In this paper, we simulate SMBHBs with damped random walk (DRW) AGN variability and an added sinusoidal signal from the orbital motion, and test three methods—a generalized Lomb–Scargle periodogram (GLSP), a nested Bayesian sampler (NBS), and a weighted wavelet z-transform (or WWZ)—to determine which is best at recovering the periodicity. Our simulated light curves follow the properties of the Catalina Real-Time Transient Survey (or CRTS), Legacy Survey of Space and Time (LSST), and Zwicky Transient Facility (ZTF) to best inform current and future SMBHB searches. We map a broad range of parameter space and identify which DRW-only light curves best mimic periodicity and pass each method’s model selection. The NBS performs best at detecting periodicity and filtering out DRW-only light curves. Combined candidate selection with both the NBS and GLSP significantly reduces false-positive rates (FPRs) with marginal impact on true-positive rates (TPRs). With this joint model selection pipeline, we find the lowest FPRs in ZTF-like simulations and the highest detection rates in LSST-like simulations. Using a modified computation of the false-alarm probability with GLSP, we efficiently triage LSST AGN light curves (∼10 7 light curves in ∼10–30 hr) and achieve TPRs and FPRs of ∼40% and ∼0.5%, respectively.

Banaszak, Sebastian M. [Vanderbilt Univ., Nashvill↗
DOE OSTI2024

Enabling the broader adoption of fusion simulation on complex geometry

This project addressed a key barrier to advanced fusion and nuclear simulation: the difficulty of performing high-fidelity Monte Carlo neutronics directly on complex, real-world CAD geometry. Traditional workflows require engineers to rebuild CAD models as simplified constructive solid geometry, a time-consuming and error-prone process that limits design iteration and broader adoption of simulation tools. The goal of this Phase I SBIR was to make CAD-based neutronics practical, accessible, and robust for industrial and research users. During the project, Coreform significantly enhanced the Direct Accelerated Geometry Monte Carlo (DAGMC) workflow and fully integrated it into Coreform Cubit as a first-class capability. Major achievements include optimized material assignment and surface meshing workflows, substantial performance improvements to geometry imprinting and preparation, native export of DAGMC models, and new visualization tools to support OpenMC source definition and lost-particle debugging. Coreform also expanded Cubit’s capabilities as a full OpenMC preprocessor, including the ability to convert OpenMC constructive solid geometry models back into CAD for visualization, multiphysics coupling, and debugging. In collaboration with Argonne National Laboratory, the project delivered comprehensive new DAGMC documentation and training materials, transforming DAGMC from a research-oriented tool into a production-ready workflow. Results were disseminated through tutorials, conference training, and multiple well-attended webinars demonstrating integrated CAD-based neutronics and multiphysics workflows. Overall, this project demonstrated that high-fidelity Monte Carlo simulations can be performed directly on complex CAD geometry, reducing setup time, improving usability, and enabling faster, more informed design decisions for fusion and nuclear energy systems.

42 ENGINEERING↗
DOE OSTI2025

Design, Build, and Initial Testing of a Portable Methane Measurement Platform

The quantification of methane concentrations in air is essential for the quantification of methane emissions, which in turn is necessary to determine absolute emissions and the efficacy of emission mitigation strategies. These are essential if countries are to meet climate goals. Large-scale deployment of methane analyzers across millions of emission sites is prohibitively expensive, and lower-cost instrumentation has been recently developed as an alternative. Currently, it is unclear how cheaper instrumentation will affect measurement resolution or accuracy. To test this, the Wireless Autonomous Transportable Methane Emission Reporting System (WATCH 4 ERS) has been developed, comprising four commercially available sensing technologies: metal oxide (MOx,), Non-dispersion Infrared (NDIR), integrated infrared (INIR), and tunable diode laser absorption spectrometer (TDLAS). WATCHERS is the accumulated knowledge of several long-term methane measurement projects at Colorado State University’s Methane Emission Technology Evaluation Center (METEC), and this study describes the integration of these sensors into a single unit and reports initial instrument response to calibration procedures and controlled release experiments. Specifically, this paper aims to describe the development of the WATCH 4 ERS unit, report initial sensor responses, and describe future research goals. Meanwhile, future work will use data gathered by multiple WATCH 4 ERS units to 1. better understand the cost–benefit balance of methane sensors, and 2. identify how decreasing instrumentation costs could increase deployment coverage and therefore inform large-scale methane monitoring strategies. Both calibration and response experiments indicate the INIR has little practical use for measuring methane concentrations less than 500 ppm. The MOx sensor is shown to have a logarithmic response to methane concentration change between background and 600 ppm but it is strongly suggested that passively sampling MOx sensors cannot respond fast enough to report concentrations that change in a sub-minute time frame. The NDIR sensor reported a linear change to methane concentration between background and 600 ppm, although there was a noticeable lag in reporting changing concentration, especially at higher values, and individual peaks could be observed throughout the experiment even when the plumes were released 5 s apart. The TDLAS sensor reported all changes in concentration but remains prohibitively expensive. Our findings suggest that each sensor technology could be optimized by either operational design or deployment location to quantify methane emissions. The WATCH4ERS units will be deployed in real-world environments to investigate the utility of each in the future.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

New Beginnings: Optimising Waste and End-state at UK Nuclear Sites to Enable Environmental Permit surrender (the GRR) - 20193

The decommissioning of nuclear sites involves the management of large volumes of radioactive waste, other conventional waste, and the clean-up of land and groundwater contamination. In the UK, the disposal of radioactive waste from nuclear sites is regulated by the Environment Agency (England), Natural Resources Wales (Wales) and the Scottish Environment Protection Agency (Scotland). All three environment agencies seek to make sure that the generation and disposal of radioactive waste is optimised: ensuring impacts to people are 'as low as reasonably achievable' (ALARA). With a large proportion of UK nuclear sites embarking on decommissioning, the environment agencies have collaboratively developed new guidance: the 'Management of radioactive waste from the decommissioning of nuclear sites: Guidance on the Requirements for Release from radioactive substances regulation' (referred to as the GRR)[1]. The environment agencies worked in liaison with the Office for Nuclear Regulation (ONR), the UK nuclear safety regulator, and the UK government body responsible, the Nuclear Decommissioning Authority (NDA). Prior to finalising the guidance, a 12 month trial was completed at three of the NDA's nuclear sites, enabling the environment agencies to test their draft GRR guidance against 'real world scenarios' and the challenge of incorporating its requirements into decommissioning plans. Experience gained during the trial was incorporated into the current version of the GRR, published in July 2018. The GRR represents the 'new beginning' of this paper's title because for the first time, the UK environmental regulators have expressed in detail their requirements for surrender of environmental permits for nuclear sites. The GRR provides a framework for a UK nuclear site to plan, assess and justify its management of radioactive waste and deliver through decommissioning, an optimal, environmentally safe and sustainable end-state. This paper will describe the requirements and technical standards within the GRR, the establishment of defensible clean-up criteria and the production of a Waste Management Plan (WMP), supported by a Site-Wide Environmental Safety Case (SWESC). It will describe the experience gleaned by operators and regulators during the trial and it will highlight current work by regulators to implement the new approach proportionately. Finally, this paper will look forward and outline planned amendments to legislation in the UK affecting nuclear sites that should enable the full benefits of the GRR to be realised in the future. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2024

High- and Mid-Fidelity Modeling Comparison for a Floating Marine Turbine System

There is a lack of suitable numerical tools, particularly open-source tools, that can be used for designing and optimizing marine turbine systems. The National Renewable Energy Laboratory has added features to their widely used mid-fidelity wind turbine modeling code, OpenFAST, to enable modeling of axial-flow marine turbines. This necessitated the addition of several physical effects relevant to marine turbines that are neglected for wind turbines. These include buoyancy, added mass and inertial loads, wave-current superposition, and changes to the coordinate systems. This updated version of OpenFAST allows for the modeling of both fixed and floating marine turbines at a speed comparable to real time. While efficient for large sets of load cases and design studies, mid-fidelity codes make simplifying assumptions that may impact their accuracy. High-fidelity computational fluid dynamics (CFD) simulations can capture more flow effects with fewer assumptions and provide detailed body pressure mapping and flow-field information. It is important to compare predictions between mid-fidelity and high-fidelity codes, both to verify the models and to understand the limitations. A floating marine turbine system was modeled both with OpenFAST and with the commercial CFD code STAR-CCM+. The CFD model used a three-dimensional unsteady Reynolds-averaged Navier-Stokes solver for a volume-of-fluid numerical wave and current tank. The blade-resolved simulations used the sliding-interface technique for the spinning rotor and an overset grid to accommodate the rigid-body motion of the floating system. The mooring system was modeled with a custom coupling of the CFD solver with the open-source code MoorDyn. This improves upon the existing quasi-static catenary solver in STAR-CCM+, which lacks seabed contact or line-to-line connections. Simulation results for a floating marine turbine are compared between OpenFAST and CFD, highlighting the capabilities of the mid-fidelity code and identifying the areas where a high-fidelity approach is needed.

CFD↗
DOE OSTI2020

Monitoring the Caustic Dissolution of Aluminum Alloy in a Radiochemical Hot Cell Using Raman Spectroscopy

Chemical processing of highly radioactive materials commonly takes place in heavily shielded hot cells. The remote, real-time monitoring of chemical processing streams via optical spectroscopic techniques in hot cells may be particularly useful. Here in this paper, we describe the implementation of Raman spectroscopy and chemometric analysis to monitor the dissolution of aluminum-clad targets containing irradiated aluminum–neptunium oxide cermet pellets in caustic solutions in a hot cell environment. Partial least squares regression analysis was used to generate calibration models to quantify the concentration of dissolved aluminum, nitrate, and hydroxide in solutions within the radiochemical hot cell. This work explored a systematic approach to optimize a matrix of calibration standards using a D-optimal experimental design. The Design of Experiments-based regression model, in comparison to more traditional analytical approaches, was found to be the more practical method for building calibration models, with fewer samples, to obtain informative analytical data from Raman spectra.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Direction-optimizing Label Propagation Framework for Structure Detection in Graphs: Design, Implementation, and Experimental Analysis

Label Propagation is not only a well-known machine learning algorithm for classification but also an effective method for discovering communities and connected components in networks. We propose a new Direction-optimizing Label Propagation Algorithm (DOLPA) framework that enhances the performance of the standard Label Propagation Algorithm (LPA), increases its scalability, and extends its versatility and application scope. As a central feature, the DOLPA framework relies on the use of frontiers and alternates between label push and label pull operations to attain high performance. It is formulated in such a way that the same basic algorithm can be used for finding communities or connected components in graphs by only changing the objective function used. Additionally, DOLPA has parameters for tuning the processing order of vertices in a graph to reduce the number of edges visited and improve the quality of solution obtained. We present the design and implementation of the enhanced algorithm as well as our shared-memory parallelization of it using OpenMP. We also present an extensive experimental evaluation of our implementations using the LFR benchmark and real-world networks drawn from various domains. Compared with an implementation of LPA for community detection available in a widely used network analysis software, we achieve at most five times the F-Score while maintaining similar runtime for graphs with overlapping communities. We also compare DOLPA against an implementation of the Louvain method for community detection using the same LFR-graphs and show that DOLPA achieves about three times the F-Score at just 10% of the runtime. For connected component decomposition, our algorithm achieves orders of magnitude speedups over the basic LP-based algorithm on large-diameter graphs, up to 13.2× speedup over the Shiloach-Vishkin algorithm, and up to 1.6× speedup over Afforest on an Intel Xeon processor using 40 threads.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Instability Issue of Paralleled Dies in an SiC Power Module in Solid-State Circuit Breaker Applications

Paralleled dies in a power module could have instability issues during high current switching transients. Here, the instability is caused by the differential-mode oscillation among paralleled MOSFETs. Conventional analyses of paralleled MOSFETs’ stability are normally limited to a single operating point, which ignores the influences of the switching trajectory and nonlinear device parameters on stability. This article reveals that the switching trajectory can significantly influence parallel stability. The analysis is improved by solving eigenvalues of state-space modeling system matrices of all operating points that the switching trajectory goes through considering nonlinear device parameters. Higher voltage and current stresses result in greater real parts of complex eigenvalues, which explains why the paralleled MOSFETs are more unstable with higher voltage and current stresses. To improve stability in solid-state circuit breaker applications, we propose a method to manipulate the switching trajectory to avoid the unstable region where the conventional hard switching trajectory normally goes through. Experimental results show that the turn- off current capability can be increased from ~five times of rated current with the gate oscillation using the conventional turn- off trajectory to ~ten times of rated current without the gate oscillation using the optimal turn- off trajectory.

42 ENGINEERING↗
DOE OSTI2020

Competing Interests and Creating Value - Balancing Durable and Enforceable Institutional Controls with Beneficial Reuse - 20320

US DOE LM has various real property interests at the 100 sites within its current portfolio, with an anticipated total of 127 sites within the next 5 years. Each site has its own narrative with respect to the real property assets. Since the regulatory time frame for long-term surveillance and maintenance activities at some sites can be in perpetuity, durable and effective instruments and documentation are necessary to support LM's unique mission as a federal land manager and steward. More than half of the sites currently in LM's inventory do not allow unrestricted use, and institutional controls (ICs) are required to limit human and environmental exposures to residual contamination. LM's Asset Management program balances protection of human health and the environment with a goal of optimizing its land portfolio through mitigation of risk, limiting liability, and creating value, whether tangible or intangible, in these types of assets. The IC layering strategy utilizes multiple types, restrictions, and legal authorities to protect human health and the environment. This layering strategy influences the transition of these properties into LM's portfolio to assess potential beneficial reuses. The LM's beneficial reuse program aims to repurpose former contaminated sites to restore the environment, protect the public health, revitalize communities, and spur economic growth. LM strives to create value in a perceived valueless asset and reevaluates and revisits the potential opportunities for these sites as long-term surveillance and maintenance activities are performed and objectives are achieved. Beneficial reuse also promotes protectiveness by ensuring activities are compatible with long-term maintenance and protection of public health and the environment as well as ensuring activities are environmentally sound by retaining good stewardship of natural resources. LM has recently revised both the Institutional Control Program Guidance and its Beneficial Reuse Management Plan to integrate the issues of land use with ICs, creating value in these impaired assets. This paper presents the following topics: (1) a general overview of LM's IC and beneficial reuse programs; (2) the unique aspects of each of these programs as a result of the multiple regulatory requirements and programs that affect LM sites; (3) case studies of LM sites to show examples of how LM created value while complying with the ICs and layering strategies; and (4) the most common challenges to beneficial reuses at LM sites and potential options for value creation in impaired assets. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2025

Edge AI-Enhanced Traffic Monitoring and Anomaly Detection Using Multimodal Large Language Models

This paper addresses the challenge of traffic monitoring and incident detection in remote areas, utilizing multimodal large language models (LLMs) deployed on edge AI devices. The key novelty of the LLM is to convert real-time video streams into descriptive texts, enabling low-bandwidth transmissions and reliable detection of anomalies and incidents in environments of intermittent connectivity. The model is developed based on fine-tuning open-source LLMs and extending it with multi-modal capabilities to analyze video frames. Our work also involves deploying this model on edge devices such as Nvidia IGX Orin and is planned to be tested in realistic environments in future work. The methodology includes data set curation, iterative model fine-tuning and compression, and hardware-based optimization. This approach aims to enhance traffic safety and response speed in remote areas, marking a significant advancement in the application of AI for traffic monitoring and safety management.

Peruski, Ryan [University of Tennessee, Knoxville ↗
DOE OSTI2025

Accelerating GNNs on GPU Sparse Tensor Cores through N:M Sparsity-Oriented Graph Reordering

Recent advancements in GPU hardware support have introduced the capability to leverage N:M sparse patterns for substantial performance gains. Graphs in Graph Neural Networks (GNNs) are typically sparse, but the sparsity is often irregular, not conforming to such sparse patterns. In this paper, we propose a novel graph reordering algorithm, the first of its kind, to reshape irregular graph data into the N:M structured sparse pattern at the tile level, allowing linear-algebra-based graph operations in GNNs to benefit from the N:M sparse hardware. The optimization is lossless, maintaining the accuracy of GNN. It can remove 98-100\% violations of the N:M sparse patterns at the vector level, and increase the proportion of conforming graphs in SuiteSparse collection from 5-9\% to 88.7-93.5\%. On A100 GPUs, the optimization accelerates Sparse Matrix Matrix (SpMM) by up to 43X (2.3X -- 7.5X on average) and speeds up the key graph operations in GNNs on real graphs by as much as 8.6X (3.5X on average).

artificial intelligence, graph neural networks↗
DOE OSTI2025

Experimental test of model predictive control in a variable air volume system

Model predictive control (MPC) has been widely studied as a promising approach for improving energy efficiency and operational flexibility in buildings, yet its real-world performance for commercial variable air volume (VAV) systems remains insufficiently characterized. In particular, the impacts of model mismatch on control robustness, real-time computational burden, and device-level operation are rarely evaluated using long-term field data. Here, this study presents a comprehensive experimental evaluation of MPC applied to a full-scale VAV system in Oak Ridge National Laboratory’s Flexible Research Platform-2 building with constant cooling/heating temperature setpoints and no occupancy. The study offers three key advantages over existing work: (1) it uses a representative building in a full-scale experimental test, capturing realistic system dynamics and complexity; (2) it evaluates a relatively sophisticated MPC formulation using two different optimization solvers (Gurobi and PSO), fully accounting for computational complexity and methodological diversity; and (3) it systematically assesses potential negative impacts on various building devices, benchmark against a well-established baseline, ASHRAE Guideline 36 (G36). To isolate zone- and air-handling-unit–level supervisory control effects, the supply fan was operated with a fixed static pressure setpoint under all strategies, and the trim-and-response static pressure reset in G36 was not enabled. Results show that MPC maintained thermal comfort while improving energy efficiency. Abrupt solar radiation variations degraded performance. Computation times ranged from ∼1 s (Gurobi) to ∼ 70 s (PSO). Compared with G36, MPC achieves 33% energy savings and reduces median reheat coil output by approximately a factor of 5–10 for a representative cooling day under matched weather conditions. However, it increases the maximum discomfort deviation from 0.5 to 1°C and results in a 32% increase in staging frequency. In addition, PSO-based MPC introduced damper oscillations, also affecting actuator longevity.

ASHRAE guideline 36↗