Tensor Contraction and Operation Minimization for Extreme Scale Computational Chemistry
Explore the source record for details and available documents.
97 MATHEMATICS AND COMPUTING↗
SEARCH · Search NASA
Results for “MATHEMATICS - CALCUS, OPERATIONAL”
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.
Nonlocal Kernel Network (NKN): a Stable and Resolution-Independent Deep Neural Network.
Neural operators have recently become popular tools for designing solution maps between function spaces in the form of neural networks. Differently from classical scientific machine learning approaches that learn parameters of a known partial differential equation (PDE) for a single instance of the input parameters at a fixed resolution, neural operators approximate the solution map of a family of PDEs [6, 7]. Despite their success, the uses of neural operators are so far restricted to relatively shallow neural networks and confined to learning hidden governing laws. In this work, we propose a novel nonlocal neural operator, which we refer to as nonlocal kernel network (NKN), that is resolution independent, characterized by deep neural networks, and capable of handling a variety of tasks such as learning governing equations and classifying images. Our NKN stems from the interpretation of the neural network as a discrete nonlocal diffusion reaction equation that, in the limit of infinite layers, is equivalent to a parabolic nonlocal equation, whose stability is analyzed via nonlocal vector calculus. The resemblance with integral forms of neural operators allows NKNs to capture long-range dependencies in the feature space, while the continuous treatment of node-to-node interactions makes NKNs resolution independent. The resemblance with neural ODEs, reinterpreted in a nonlocal sense, and the stable network dynamics between layers allow for generalization of NKN’s optimal parameters from shallow to deep networks. This fact enables the use of shallow-to-deep initialization techniques [8]. Our tests show that NKNs outperform baseline methods in both learning governing equations and image classification tasks and generalize well to different resolutions and depths.
97 MATHEMATICS AND COMPUTING↗
Precursor Analysis Report: Doppelpaymer Ransomware Attack on Petroleos Mexicanos (PEMEX) 2019
The DoppelPaymer Ransomware Attack on Petroleos Mexicanos (PEMEX) 2019 Precursor Analysis Report leverages publicly available information about the PEMEX cyber attack and catalogs anomalous observables for each technique employed in the attack. This analysis is based upon the methodology of the Cybersecurity for the Operational Technology Environment (CyOTE) program. The 2019 DoppelPaymer ransomware attack on PEMEX, Mexico’s nationalized petroleum corporation, highlights a unique threat that ransomware and cybercriminal extortion poses to Operational Technology (OT) environments in critical infrastructure. The incident began with an employee downloading commodity malware that allowed adversaries to gain initial access to PEMEX’s enterprise environment. After conducting privilege escalation, tool ingress, and data exfiltration, the adversaries deployed DoppelPaymer ransomware throughout the PEMEX enterprise environment, resulting in the company having to take dozens of systems offline for at least several days. Although PEMEX stated that their operations were not affected, the data exfiltrated from PEMEX was made available for download on DoppelPaymer’s leak site, as well as on other illicit criminal forums. This stolen data included not only company information, but also sensitive OT-specific configuration data. This incident showcases how cybercriminal exfiltration and posting of sensitive OT architecture documentation can pose security concerns for the targeted organization for years due to the long lifespan of OT assets and architectures. Researchers and analysts identified 18 unique techniques utilized during the attack with a total of 190 observables using MITRE ATT&CK® for Industrial Control Systems. The CyOTE program assesses observables accompanying techniques used prior to the triggering event to identify opportunities to detect malicious activity. If observables accompanying the attack techniques are perceived and investigated prior to the triggering event, earlier comprehension of malicious activity can take place. Fifteen of the identified techniques used during the DoppelPaymer ransomware attack were precursors to the triggering event. Analysis identified 163 observables associated with these precursor techniques, 34 of which were assessed to have an increased likelihood of being perceived in the 60 days preceding the triggering event. The response and comprehension time could have been reduced if the observables had been identified earlier. The information gathered in this report contributes to a library of observables tied to a repository of artifacts, data sources, and technique detection references for practitioners and developers to support the comprehension of indicators of attack. Asset owners and operators can use these products if they experience similar observables or to prepare for comparable scenarios.
45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
Developing the Proof of Concept for the SERI QC Flag Translation
SERI QC is a robust solar data quality assessment software tool that has been in continuous use for more than three decades. This report, the second of six in the Data Quality and Uncertainty Integration project, develops the proof of concept, including programming in C and a collection of spreadsheet tools, for estimating the operational uncertainty from three-component solar irradiance measurements using information available from SERI QC. The report establishes a field reference for determining the operational uncertainty for each irradiance component based on the SERI QC coupling. Results demonstrate that the operational uncertainty estimates based on global irradiance best represent the collective results for all sky conditions.
14 SOLAR ENERGY↗
Mitigating spectral bias in neural operators via high-frequency scaling for physical systems
Neural operators have emerged as powerful surrogates for modeling complex physical problems. However, they suffer from spectral bias making them oblivious to high-frequency modes, which are present in multiscale physical systems. Therefore, they tend to produce over-smoothed solutions, which is particularly problematic in modeling turbulence and for systems with intricate patterns and sharp gradients such as multi-phase flow systems. In this work, we introduce a new approach named high-frequency scaling (HFS) to mitigate spectral bias in convolutional-based neural operators. By integrating HFS with proper variants of UNet, we demonstrate a higher prediction accuracy by mitigating spectral bias in single and two-phase flow problems. Unlike Fourierbased techniques, HFS is directly applied to the latent space, thus eliminating the computational cost associated with the Fourier transform. Additionally, we investigate alternative spectral bias mitigation through a diffusion model conditioned on neural operators. While the diffusion model integrated with the standard neural operator may still suffer from significant errors, these errors are substantially reduced when the diffusion model is integrated with a HFS-enhanced neural operator.
97 MATHEMATICS AND COMPUTING↗
New Technique of High-Performance Torque Control Developed for Induction Machines
Two forms of high-performance torque control for motor drives have been described in the literature: field orientation control and direct torque control. Field orientation control has been the method of choice for previous NASA electromechanical actuator research efforts with induction motors. Direct torque control has the potential to offer some advantages over field orientation, including ease of implementation and faster response. However, the most common form of direct torque control is not suitable for the highspeed, low-stator-flux linkage induction machines designed for electromechanical actuators with the presently available sample rates of digital control systems (higher sample rates are required). In addition, this form of direct torque control is not suitable for the addition of a high-frequency carrier signal necessary for the "self-sensing" (sensorless) position estimation technique. This technique enables low- and zero-speed position sensorless operation of the machine. Sensorless operation is desirable to reduce the number of necessary feedback signals and transducers, thus improving the reliability and reducing the mass and volume of the system. This research was directed at developing an alternative form of direct torque control known as a "deadbeat," or inverse model, solution. This form uses pulse-width modulation of the voltage applied to the machine, thus reducing the necessary sample and switching frequency for the high-speed NASA motor. In addition, the structure of the deadbeat form allows the addition of the high-frequency carrier signal so that low- and zero-speed sensorless operation is possible. The new deadbeat solution is based on using the stator and rotor flux as state variables. This choice of state variables leads to a simple graphical representation of the solution as the intersection of a constant torque line with a constant stator flux circle. Previous solutions have been expressed only in complex mathematical terms without a method to clearly visualize the solution. The graphical technique allows a more insightful understanding of the operation of the machine under various conditions.
Kenny, Barbara H.↗
Field Insights: Strengthening Digital Assurance Through On-Site Network Monitoring
The accelerating deployment of digital energy infrastructure, ranging from inverter-based resources (IBRs), battery energy storage systems (BESS), to advanced grid control platforms, has brought unprecedented visibility, flexibility, and efficiency to the electric grid. However, this digital transformation also introduces new cybersecurity challenges, particularly in the form of supply chain risks and operational blind spots at the grid edge. Over the past year, the Department of Energy’s Office of Cybersecurity, Energy Security, and Emergency Response (CESER), through its Rapid Risk Assessment initiative, along with the Grid Deployment Office (GDO), through its Technical Assistance for Digital Assurance (TADA) initiative, have supported a series of on-site network engagements led by Idaho National Laboratory (INL). These engagements, conducted in partnership with asset owners across the country, have focused on identifying real-world vulnerabilities and misconfigurations in operational environments, many of which are not detectable through remote assessments or traditional compliance audits. The goal of this report is to distill key findings and lessons learned during network hunt engagements from INL’s fiscal year (FY) 2024 - 2025. It is intended to help asset owners—regardless of their participation in the program—better understand the evolving threat landscape and adopt practical measures to secure their digital energy infrastructure.
24 - POWER TRANSMISSION AND DISTRIBUTION↗
Solutions of some problems in applied mathematics using MACSYMA
Various Symbolic Manipulation Programs (SMP) were tested to check the functioning of their commands and suitability under various operating systems. Support systems for SMP were found to be relatively better than the one for MACSYMA. The graphics facilities for MACSYMA do not work as expected under the UNIX operating system. Not all commands for MACSYMA function as described in the manuals. Shape representation is a central issue in computer graphics and computer-aided design. Aside from appearance, there are other application dependent, desirable properties like continuity to certain order, symmetry, axis-independence, and variation-diminishing properties. Several shape representations are studied, which include the Osculatory Method, a Piecewise Cubic Polynomial Method using two different slope estimates, Piecewise Cubic Hermite Form, a method by Harry McLaughlin, and a Piecewise Bezier Method. They are applied to collected physical and chemical data. Relative merits and demerits of these methods are examined. Kinematics of a single link, non-dissipative robot arm is studied using MACSYMA. Lagranian is set-up and Lagrange's equations are derived. From there, Hamiltonian equations of motion are obtained. Equations suggest that bifurcation of solutions can occur, depending upon the value of a single parameter. Using the characteristic function W, the Hamilton-Jacobi equation is derived. It is shown that the H-J equation can be solved in closed form. Analytical solutions to the H-J equation are obtained.
Punjabi, Alkesh↗
State-of-the-art of data collection, analytics, and future needs of transmission utilities worldwide to account for the continuous growth of sensing data
Nowadays, transmission system operators require higher degree of observability in real-time to gain situational awareness and improve the decision-making process to guarantee a safe and reliable operation. Digitalization of energy systems allows utilities to monitor the system dynamic performance in real-time at fast time scales. The use of such technologies has unlocked new opportunities to introduce new data driven algorithms for improving the stability assessment and control of the system. Motivated by these challenges, a group of experts have worked together to highlight and establish a baseline set of these common concerns, which can be used as motivation to propose innovative analytics and data-driven solutions. In this document, the results of a survey on 10 transmission system operators around the world are presented and it aims to understand the current practices of the participating companies, in terms of data acquisition, handling, storage, modelling and analytics. The overall objective of this document is to capture the actual needs from the interviewed utilities, thereby laying the groundwork for setting valid assumptions for the development of advanced algorithms in this field.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Co-Simulation Meets AI: MCP-Driven Power System Analysis
GridGPT, a fine-tuned Generative AI model is designed for on-premise use in grid control rooms. This presentation will demonstrate how eGridGPT can seamlessly integrate with control room solutions to offer operators, engineers, and corporate users enhanced guidance and decision support. It is to show how this innovative AI solution can improve state estimation, boost variable energy forecasting, and optimize grid operations. By leveraging eGridGPT's unique features, audience will learn to unlock new levels of automation, predictive analytics, and reliability within their power systems, ultimately leading to reduced downtime and improved operational efficiency.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Exploring Advanced Computational Tools and Techniques with Artificial Intelligence and Machine Learning in Operating Nuclear Plants
This report presents the project Idaho National Laboratory conducted for Nuclear Regulatory Commission to explore the advanced computational tools and techniques, such as artificial intelligence (AI) and machine learning (ML), for operating nuclear plants. The report reviews the nuclear data sources, with the focus on the operating experience data, that could be applied by advanced computational tools and techniques. Plant-specific and generic (national and international) data from different sources are described. The report describes the relationships between statistics and AI/ML and then introduces the most widely used AI/ML algorithms in both supervised and unsupervised learning. The report reviews the recent applications of advanced computational tools and techniques in various fields of nuclear industry, such as reactor system design and analysis, plant operation and maintenance, and nuclear safety and risk analysis. Finally, the report presents the insights from the project on the potential applicability of AI/ML techniques in improving advanced computational capabilities, how the advanced tools and techniques could contribute to the understanding of safety and risk, and what information would be needed to provide meaningful insights to decision makers. The report also documents an NRC survey on the current state of commercial nuclear power operations relative to the use of AI and ML tools as well as the role of AI/ML tools in nuclear power operations was published by the NRC as in FRN NRC-2021-0048 in April 2021. A summary of the survey including the survey questions, survey participants, survey responses, and the conclusions and insights derived from the survey is provided in the report. Finally, the report investigates potential applications of using AI/ML in operating NPPs and advanced reactors (both advanced LWRs and advanced NLWRs) to improve nuclear plant safety and efficiency. Three main application fields are defined and discussed: (1) plant safety and security assessments; (2) plant degradation modeling, fault and accident diagnosis and prognosis; and (3) plant operation and maintenance efficiency improvement.
22 GENERAL STUDIES OF NUCLEAR REACTORS↗
Anomaly Detection in Test Equipment via Sliding Mode Observers
Nonlinear observers were originally developed based on the ideas of variable structure control, and for the purpose of detecting disturbances in complex systems. In this anomaly detection application, these observers were designed for estimating the distributed state of fluid flow in a pipe described by a class of advection equations. The observer algorithm uses collected data in a piping system to estimate the distributed system state (pressure and velocity along a pipe containing liquid gas propellant flow) using only boundary measurements. These estimates are then used to further estimate and localize possible anomalies such as leaks or foreign objects, and instrumentation metering problems such as incorrect flow meter orifice plate size. The observer algorithm has the following parts: a mathematical model of the fluid flow, observer control algorithm, and an anomaly identification algorithm. The main functional operation of the algorithm is in creating the sliding mode in the observer system implemented as software. Once the sliding mode starts in the system, the equivalent value of the discontinuous function in sliding mode can be obtained by filtering out the high-frequency chattering component. In control theory, "observers" are dynamic algorithms for the online estimation of the current state of a dynamic system by measurements of an output of the system. Classical linear observers can provide optimal estimates of a system state in case of uncertainty modeled by white noise. For nonlinear cases, the theory of nonlinear observers has been developed and its success is mainly due to the sliding mode approach. Using the mathematical theory of variable structure systems with sliding modes, the observer algorithm is designed in such a way that it steers the output of the model to the output of the system obtained via a variety of sensors, in spite of possible mismatches between the assumed model and actual system. The unique properties of sliding mode control allow not only control of the model internal states to the states of the real-life system, but also identification of the disturbance or anomaly that may occur.
Solano, Wanda M.↗
Modeling and Validation of a Propellant Mixer for Controller Design
A mixing chamber used in rocket engine testing at the NASA Stennis Space Center is modelled by a system of two nonlinear ordinary differential equations. The mixer is used to condition the thermodynamic properties of cryogenic liquid propellant by controlled injection of the same substance in the gaseous phase. The three inputs of the mixer are the positions of the valves regulating the liquid and gas flows at the inlets, and the position of the exit valve regulating the flow of conditioned propellant. Mixer operation during a test requires the regulation of its internal pressure, exit mass flow, and exit temperature. A mathematical model is developed to facilitate subsequent controller designs. The model must be simple enough to lend itself to subsequent feedback controller design, yet its accuracy must be tested against real data. For this reason, the model includes function calls to thermodynamic property data. Some structural properties of the resulting model that pertain to controller design, such as uniqueness of the equilibrium point, feedback linearizability and local stability are shown to hold under conditions having direct physical interpretation. The existence of fixed valve positions that attain a desired operating condition is also shown. Validation of the model against real data is likewise provided.
Richter, Hanz↗
Analysis of the Capacity Potential of Current Day and Novel Configurations for New York's John F. Kennedy Airport
In 2015, a series of systems analysis studies were conducted on John F. Kennedy Airport in New York (NY) in a collaborative effort between NASA and the Port Authority of New York and New Jersey (PANYNJ). This work was performed to build a deeper understanding of NY airspace and operations to determine the improvements possible through operational changes with tools currently available, and where new technology is required for additional improvement. The analysis was conducted using tool-based mathematical analyses, video inspection and evaluation using recorded arrival/departure/surface traffic captured by the Aerobahn tool (used by Kennedy Airport for surface metering), and aural data archives available publically through the web to inform the video segments. A discussion of impacts of trajectory and operational choices on capacity is presented, including runway configuration and usage (parallel, converging, crossing, shared, independent, staggered), arrival and departure route characteristics (fix sharing, merges, splits), and how compression of traffic is staged. The authorization in March of 2015 for New York to use reduced spacing under the Federal Aviation Administration (FAA) Wake Turbulence Recategorization (RECAT) also offers significant capacity benefit for New York airports when fully transitioned to the new spacing requirements, and the impact of these changes for New York is discussed. Arrival and departure capacity results are presented for each of the current day Kennedy Airport configurations. While the tools allow many variations of user-selected conditions, the analysis for these studies used arrival-priority, no-winds, additional safety buffer of 5% to the required minimum spacing, and a mix of traffic typical for Kennedy. Two additional "novel" configurations were evaluated. These configurations are of interest to Port Authority and to their airline customers, and are believed to offer near-term capacity benefit with minimal operational and equipage changes. One of these is the addition of an Optimized Profile Descent (OPD) route to runways 22L and 22R, and the other is the simultaneous use of 4 runways, which is not currently done at Kennedy. The background and configuration for each of these is described, and the capacity results are presented along with a discussion of drawbacks and enablers for each.
Glaab, Patricia↗
A novel machine learning-based optimization algorithm (ActivO) for accelerating simulation-driven engine design
A novel design optimization approach (ActivO) that employs an ensemble of machine learning algorithms is presented. The proposed approach is a surrogate-based scheme, where the predictions of a weak leaner and a strong learner are utilized within an active learning loop. The weak learner is used to identify promising regions within the design space to explore, while the strong learner is used to determine the exact location of the optimum within promising regions. For each design iteration, exploration is done by randomly selecting evaluation points within regions where the weak learner-predicted fitness is high. The global optimum obtained by using the strong learner as a surrogate is also evaluated to enable rapid convergence once the most promising region has been identified. First, the performance of ActivO was compared against five other optimizers on a cosine mixture function with 25 local optima and one global optimum. In the second problem, the objective was to minimize indicated specific fuel consumption of a compression-ignition internal combustion (IC) engine while adhering to desired constraints associated with in-cylinder pressure and emissions. In this work, the efficacy of the proposed approach is compared to that of a genetic algorithm, which is widely used within the internal combustion engine community for engine optimization, showing that ActivO reduces the number of function evaluations needed to reach the global optimum, and thereby time-to-design by 80%. Furthermore, the optimization of engine design parameters leads to savings of around 1.9% in energy consumption, while maintaining operability and acceptable pollutant emissions.
97 MATHEMATICS AND COMPUTING↗
Not so HOT Triangulations
Here, we propose primal–dual mesh optimization algorithms that overcome shortcomings of the standard algorithm while retaining some of its desirable features. “Hodge-Optimized Triangulations” defines the “HOT energy” as a bound on the discretization error of the diagonalized Delaunay Hodge star operator. HOT energy is a natural choice for an objective function, but unstable for both mathematical and algorithmic reasons: it has minima for collapsed edges, and its extrapolation to non-regular triangulations is inaccurate and has unbounded minima. We propose a different extrapolation with a stronger theoretical foundation, and avoid extrapolation by recalculating the objective just beyond the flip threshold. We propose new objectives, based on normalizations of the HOT energy, with barriers to edge collapses and other undesirable configurations. We propose mesh improvement algorithms coupling these. When HOT optimization nearly collapses an edge, we actually collapse the edge. Otherwise, we use the barrier objective to update positions and weights and remove vertices. By combining discrete connectivity changes with continuous optimization, we more fully explore the space of possible meshes and obtain higher quality solutions.
97 MATHEMATICS AND COMPUTING↗
A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder
Traditional linear subspace reduced order models (LS-ROMs) are able to accelerate physical simulations in which the intrinsic solution space falls into a subspace with a small dimension, i.e., the solution space has a small Kolmogorov n-width. However, for physical phenomena not of this type, e.g., any advection-dominated flow phenomena such as in traffic flow, atmospheric flows, and air flow over vehicles, a low-dimensional linear subspace poorly approximates the solution. To address cases such as these, we have developed a fast and accurate physics-informed neural network ROM, namely nonlinear manifold ROM (NM-ROM), which can better approximate high-fidelity model solutions with a smaller latent space dimension than the LS-ROMs. Our method takes advantage of the existing numerical methods that are used to solve the corresponding full order models. The efficiency is achieved by developing a hyper-reduction technique in the context of the NM-ROM. Numerical results show that neural networks can learn a more efficient latent space representation on advection-dominated data from 1D and 2D Burgers' equations. A speedup of up to 2.6 for 1D Burgers' and a speedup of 11.7 for 2D Burgers' equations are achieved with an appropriate treatment of the nonlinear terms through a hyper-reduction technique. Lastly, a posteriori error bounds for the NM-ROMs are derived that take account of the hyper-reduced operators.
97 MATHEMATICS AND COMPUTING↗
HPC Forecast: Cloudy and Uncertain
An examination of how the technology landscape has changed and possible future directions for HPC operations and innovation.
97 MATHEMATICS AND COMPUTING↗