Graph Neural Network (GNN) for Property Prediction
Presented at the DOE-NETL 2020 Virtual Meeting - SMART Initiative Annual Review Meeting November 2-3, 2020, 11/2
SEARCH · Search NASA
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.
Presented at the DOE-NETL 2020 Virtual Meeting - SMART Initiative Annual Review Meeting November 2-3, 2020, 11/2
A solution methodology and implementation components are presented that can uncover unwanted, unintentional or unanticipated effects on electric grids from changes to actual electric grid control software. A new design is presented to leapfrog over the limitations of current modeling and testing techniques for cyber technologies in electric grids. We design a fully virtualized approach in which actual, unmodified operational software under test is enabled to interact with simulated surrogates of electric grids. It enables the software to influence the (simulated) grid operation and vice versa in a controlled, high fidelity environment. Challenges in achieving such capability include achieving low-overhead time control mechanisms in hypervisor schedulers, network capture and time-stamping, translation of network packets emanating from grid software into discrete events of virtual grid models, translation back from virtual sensors/actuators into data packets to control software, and transplanting the entire system onto an accurately and efficiently maintained virtual-time plane.
The rapid growth of the Internet of Things (IoT) and Edge Computing (EC) has brought significant conveniences to modern society but has also greatly expanded the cyber attack surfaces, particularly as these technologies are being increasingly integrated into critical systems such as power grids, healthcare, and smart homes. Here, to improve IoT/EC’s cybersecurity posture, we leveraged Artificial Intelligence (AI) and Machine Learning (ML) by employing tinyML to monitor voluminous IoT data for cyber threats while addressing devices’ resource constraints, and utilizing Federated Learning (FL) to share local detection knowledge across the system while preserving privacy. Building on our three-layer architecture combining tinyML and FL to enhance autonomous cyber attack detection, this paper demonstrated that the architecture improves detection accuracy, reduces resource consumption, and enables lightweight, secure IoT device monitoring. These results were validated using the public N-BaIoT dataset as well as real IoT network traffic data collected under multiple attack scenarios from our testbeds. Additionally, we introduced an enhanced FL methodology with a novel preprocessing stage, including federated feature selection and global preprocessor construction, to address IoT/EC data heterogeneity. We developed a physical IoT testbed for attack simulations and data collection, implemented a tinyML-powered detector for realistic model validation, and also built a virtual testbed for scalable evaluations of FL models across diverse network environments.
Abstract Multispectral imaging has been used for numerous applications in e.g., environmental monitoring, aerospace, defense, and biomedicine. Here, we present a diffractive optical network-based multispectral imaging system trained using deep learning to create a virtual spectral filter array at the output image field-of-view. This diffractive multispectral imager performs spatially-coherent imaging over a large spectrum, and at the same time, routes a pre-determined set of spectral channels onto an array of pixels at the output plane, converting a monochrome focal-plane array or image sensor into a multispectral imaging device without any spectral filters or image recovery algorithms. Furthermore, the spectral responsivity of this diffractive multispectral imager is not sensitive to input polarization states. Through numerical simulations, we present different diffractive network designs that achieve snapshot multispectral imaging with 4, 9 and 16 unique spectral bands within the visible spectrum, based on passive spatially-structured diffractive surfaces, with a compact design that axially spans ~72 λ m , where λ m is the mean wavelength of the spectral band of interest. Moreover, we experimentally demonstrate a diffractive multispectral imager based on a 3D-printed diffractive network that creates at its output image plane a spatially repeating virtual spectral filter array with 2 × 2 = 4 unique bands at terahertz spectrum. Due to their compact form factor and computation-free, power-efficient and polarization-insensitive forward operation, diffractive multispectral imagers can be transformative for various imaging and sensing applications and be used at different parts of the electromagnetic spectrum where high-density and wide-area multispectral pixel arrays are not widely available.
Data-intensive applications, including high energy and nuclear physics, astrophysics, climate modeling, nano-scale materials science, genomics, and financing, are expected to generate exabytes of data over the coming years, which must be transferred, visualized, and analyzed by geographically distributed teams of users. High-performance network capabilities must be available to these users at the application level in a transparent, virtualized manner. Moreover, the application users must have the capability to move large datasets from local and remote locations across network environments to their home institutions. To solve these challenges, the main goal of our project is to design and evaluate high-performance data transfer software to support various data-intensive applications. First, we have designed a middleware software that provides access to Remote Direct Memory Access (RDMA) functionalities. This middleware integrates network access, memory management and multitasking in its core design. We address a number of issues related to its efficient implementation, for instance, explicit buffer management and memory registration, and parallelization of RDMA operations, which are vital to delivering the benefit of RDMA to the applications. Built on top of this middleware, an implementation and experimental evaluation of the RDMA-based FTP software, RFTP, is described and evaluated. This application has been implemented by our team to exploit the full capabilities of advanced RDMA mechanisms for ultra-high speed bulk data transfer applications on Energy Sciences Network (ESnet). Second, we designed our data transfer software to optimize TCP/IP based data transfer performance such that RFTP can be fully compatible with today’s Internet. Our kernel optimization techniques with Linux system calls sendfile and splice, can reduce data copy cost. In this report, we summarize the technical challenges of our project, the primary software design methods, the major project milestones achieved, as well as the testbed evaluation work and demonstrations during our project life time.
We propose a new method for solving an important problem of astronomy that arises in observations with ultrahigh-angular-resolution interferometers. This method is based on the application of the theory of artificial neural networks. We propose and compute a multiparameter model for a celestial object like Sgr A*. For this model we have numerically constructed a number of probable images for neural network training. After neural network training on these images, the quality of its operation has been tested on another series of images from the same model. We have proven that a neural network can recognize and classify celestial objects (also obtained from interferometers) virtually no worse than can be done by a human.
Nuclear detectors for x-ray and gamma-ray spectroscopy and imaging are a vital tool in many homeland security, medical imaging, astrophysics and other applications. Most of these applications require room-temperature operation due to the operational constraints imposed by a cryogenic cooling system. CdZnTe (CZT) has been the main material with the desired detection properties, and CZT crystals have been used commercially for three decades. However, CdZnTe still suffers from long-standing issues of high densities of performance-limiting intrinsic defects such as Te inclusions and networks of dislocation walls (sub-grain boundaries). A recently invented new quaternary material CdZnTeSe showed excellent material properties for radiation detection. The material was found to be free from dislocation networks, possess reduced Te inclusions, and have better compositional homogeneity. Virtual Frisch grid detectors were fabricated from crystals taken from a CdZnTeSe ingot that was grown by the traveling heater method. The detectors were fabricated from an as-grown ingot, bypassing the post-growth annealing process commonly practiced for industrial-grade CZT. The performances of the detectors were studied with different Frisch grid lengths using an amplifier shaping time ranging from 1–6 µs. The detectors showed high-quality spectroscopic performance with an as-measured energy resolution of ~1.1% at 662 keV for an optimum Frisch grid length of 3 mm. The charge collection was observed to enhance for longer Frisch grids.
Traffic congestion is a serious problem in the US, and traffic signal control is one of the effective solutions to congestion. Previous research on model predictive control (MPC)-based traffic signal control showed substantial benefits over conventional methods. This study focused on implementing MPC over a large-scale network with complex intersections and the impact of cycle length, network size, and imperfect state estimation on performances. This study implemented a virtual phase link (VPL)-based model predictive control method which used the number of vehicles in each VPL as input state variables and was suitable for National Electrical Manufacturing Association (NEMA) ring-barrier control. To test the impact of network size, the performance of distributed MPC (36 intersections in the network are divided into five subnetworks) was compared with that of MPC over the full network for a set of cycle lengths. To test the impact of imperfect state estimation, we synthetically infused estimation error and developed two scenarios, MPC-error and MPC-error narrow, which had higher and lower estimation errors, respectively. The performance of these MPC methods was compared with that of the existing time-of-day (TOD) method and an offline method that used Webster's method for split and MULTIBAND for cycle length and offset optimization. Trajectory and linkwise signal performance measures were collected from the simulation to evaluate performance. The distributed MPC method with perfect state estimation had the lowest delay and highest energy efficiency of all the methods. The performance of MPC decreased as the prediction inaccuracy increased. MPC-error had 7% and 11% more delay than MPC-error narrow in the morning and evening peaks, respectively. Overall, simulation results suggest that even with imperfect state estimation, MPC methods will outperform offline methods significantly.
Precise timing and data transmission within stringent time constraints are critical for numerous applications, such as robotics, virtual and augmented reality, industrial automation, energy and medical business, and various other sectors. Time-sensitive networks (TSN) and fifth-generation wireless communications (5G) are crucial for industrial communications, enabling convergent communication for various services using a common network core. Applications that are time-sensitive and require deterministic communications with low latency fall into this category, such as generation plant systems. This paper presents a simulated model of 5G-TSN for a fossil power plant based on the wired network parameters implemented in situ. Metrics analysis, comparison and future work are presented. © 2024 IEEE.
This whitepaper provides strategic insights and recommendations into security cloud-based solutions for electric utilities, encompassing operational technology (OT), virtual power plants (VPP), distributed energy resources (DERs), applications, networks, and data storage as they transition to and leverage cloud infrastructure through managed service providers (MSPs) and cloud service providers (CSPs). Principles derived from established frameworks serve as a foundation for best practices across cybersecurity projects and remove the constraints of settling on a single framework. For organizations that prefer not to integrate a specific framework altogether, elements of the proposed approach could be adopted or tailored to best fit defined requirements and expected functionalities. The Cirrus assessment, a utility cloud feasibility tool, and the roadmap it provides serve as a precursor to this paper, which seeks to be a valuable resource for defining next steps following cloud technology integration feasibility appraisal. With its comprehensive approach to adoption, the Cirrus framework offers strategic guidance on responsibly preparing for or deploying a utility cloud solution. The previously published whitepaper, “Use Case-Informed Framework for Utility Cloud Migration,” details the guiding strategy, research, and deployment of cloud solutions within electric and interconnected grid systems. Before implementing the controls suggested in this document, it is recommended that stakeholders complete Cirrus's cloud integration assessment and pair the results with their unique cybersecurity controls to form a comprehensive cloud-based utility cybersecurity plan. The Cirrus outcome will consider a series of future architectures for the grid before and after the energy transition and evaluate the arguments for and against cloud applications for each electric and interconnected grid layer. This document is a companion to the original whitepaper, "Use Case-Informed Framework for Utility Cloud Migration" to further identify and recommend security controls based on Cirrus’s cloud integration assessment output. The following whitepaper outlines the cybersecurity controls that secure cloud-service models pertinent to the electric sector using the predefined categories identify, protect, detect, and respond and recover. The objective is to outline prescriptive security controls based on the type of architecture and data stored in the cloud. The focus includes dissecting the shared responsibility model and elucidating what on-premises Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS) entail. A pivotal consideration in this context is allocating responsibility for foundational cybersecurity aspects—having used Cirrus for the cloud integration assessment. The ensuing controls detailed herein also represent a checklist of controls necessary for a secure cloud transition, equipping utilities with the knowledge to navigate this digital transformation with confidence and strategic foresight in a safe and responsible manner.
Residential and commercial buildings in the United States consume 73% of all electricity generated. Demand charges (i.e. “peak usage”) can account for 30-70% of a utility bill and continue to grow even as energy prices are decreasing, due largely in part to adoption of intermittent renewables and less efficient, more costly generation assets which must be brought online during peak demand. In order to quantitatively assess business value and either enhance existing or propose new programs for utilities, the ability to detect, generate, and simulate a building energy model has been applied to a 1,385 km2 utility district involving over 160,000 buildings in a smart grid backed by a fiber optic network. The organization operating the electrical distribution and fiber optic networks has prioritized and guided the creation of a virtual utility in order to assess a prioritized list of potential revenue streams and determine the most cost-effective programs for their organization and their rate-payers. We provide an overview of the technical capabilities developed for generally-applicable urban-scale building energy modeling to any geographical area and the utility-defined use cases being assessed.
The rise and fall of artificial neural networks is well documented in the scientific literature of both the fields of computer science and computational chemistry. Yet almost two decades later, we are now seeing a resurgence of interest in deep learning, a machine learning algorithm based on “deep” neural networks. Within the last few years, we have seen the transformative impact of deep learning the computer science domain, notably in speech recognition and computer vision, to the extent that the majority of practitioners in those field are now regularly eschewing prior established models in favor of deep learning models. In this review, we provide an introductory overview into the theory of deep neural networks and their unique properties as compared to traditional machine learning algorithms used in cheminformatics. By providing an overview of the variety of emerging applications of deep neural networks, we highlight its ubiquity and broad applicability to a wide range of challenges in the field, including QSAR, virtual screening, protein structure modeling, QM calculations, materials synthesis and property prediction. In reviewing the performance of deep neural networks, we observed a consistent outperformance against non neural networks state-of-the-art models across disparate research topics, and deep neural network based models often exceeded the “glass ceiling” expectations of their respective tasks. Coupled with the maturity of GPU-accelerated computing for training deep neural networks and the exponential growth of chemical data on which to train these networks on, we anticipate that deep learning algorithms will be a useful tool and may grow into a pivotal role for various challenges in the computational chemistry field.
Watch this video to learn how NREL researchers are evaluating the interdependencies between power systems and network communication flows when cyber and physical threats to the grid occur. NREL’s cybersecurity researchers have developed an emulated, multilayer grid environment at the ESIF, where they can safely launch attacks in the virtual world on both emulated and actual physical devices throughout the facility—and evaluate how they would respond to an attack in the real world.
Abstract Convolutional neural networks (CNNs) have been successfully used in many applications where important information about data is embedded in the order of features, such as speech and imaging. However, most tabular data do not assume a spatial relationship between features, and thus are unsuitable for modeling using CNNs. To meet this challenge, we develop a novel algorithm, image generator for tabular data (IGTD), to transform tabular data into images by assigning features to pixel positions so that similar features are close to each other in the image. The algorithm searches for an optimized assignment by minimizing the difference between the ranking of distances between features and the ranking of distances between their assigned pixels in the image. We apply IGTD to transform gene expression profiles of cancer cell lines (CCLs) and molecular descriptors of drugs into their respective image representations. Compared with existing transformation methods, IGTD generates compact image representations with better preservation of feature neighborhood structure. Evaluated on benchmark drug screening datasets, CNNs trained on IGTD image representations of CCLs and drugs exhibit a better performance of predicting anti-cancer drug response than both CNNs trained on alternative image representations and prediction models trained on the original tabular data.
Purpose: A national Australian inter-university medical physics (MP) group was formed in 2011/12, supported by Department of Health Better Access to Radiation Oncology BARO) seed funding. Core membership includes the six universities providing postgraduate MP courses. Objectives include increasing capacity, development and efficiency of national academic MP structures/systems and hence supporting education, clinical training and research, for the MP workforce support. Although the BARO scheme focuses on Radiation Oncology, the group has wider MP interests. Methods: Two further BARO seed grants were achieved: 1) for networked academic activities, including shared-resource teaching, eg using virtual reality systems; MP outreach to schools and undergraduates; developing web-based student and registrar education/resources, etc.; and 2) for conjoint ‘translational research’ posts between universities and partner hospitals, to clinically progress advanced RT technologies and to support students and registrars. Each university received 0.5 FTE post from each grant over 2 years (total: $1.75M) and leveraged local additional partner funds. Results: Total funding: $4–5M. Overall there have been 35 (mainly overseas) postholders bringing specific expertise, beginning in early 2013. Periods in Australia have been from 0.25–2 years (median=1). As well as the education activities, research projects include lung/spine SBRT, 4D RT, FFF beams, technology assessment, complex treatment planning, imaging for radiation oncology, DIR, adaptive breast, datamining, radiomics,etc. Observed positive impacts include: increased interest in MP courses, training support, translational research infrastructure and/or clinical practice in the hospitals involved, plus increased collaboration and effectiveness between the universities. Posts are continuing beyond grant end using leveraged funds, providing the basis for sustainability of some posts. Conclusion: The BARO-funded projects have cost-effectively produced a range of positive impacts on training, research and practice in hospitals and between universities. The evaluation of the specific post roles and activities, and their outcomes, has produced focused recommendations on continuation and sustainability. Funding was from the Australian federal Department of Health; leveraged funding was from partner universities and hospitals.
A greater understanding of human dynamics as they play out in both physical space and through inter-personal communication is vital for the design and development of intelligent and resilient cities. Physical context provides insight in to the space-time distribution of population and their activity patterns, while inter-personal communication can, for the first time, be measured at scale through virtual interactions. The confluence of these two enables spatial social network analysis in the built environment for edge computing, knowledge discovery, and artificial intelligence activities. In this work, we propose a novel method to discover such dynamics. We use a dataset of 72 million tweets to develop a spatially embedded network of communication, and then use community detection algorithms to explore regional and urban delineation in the United States. We find that the broad spatial delineation of communities and community sub-structure is consistent with major United States socio-cultural regions, states, and major metropolitan areas. We describe how these methods could be extended to generate a measure of socially cohesive regions that can be consistently applied anywhere there is a sufficiently rich data source. We hope the proposed methods will generate new perspectives on smart and resilient cities.
All magnetically confined plasma fusion power plant concepts (Tokamak, Spherical Tokamak, Stellarator, Mirror, ...) must exhaust the heat and plasma from the core confinement region to the material walls. The primary channel for this exhaust is through a plasma divertor which directs plasma along open magnetic field lines to a material target. The Material Plasma Exposure eXperiment (MPEX) illustrated in Figure 1, is a high-power, steady-state linear plasma device designed to produce the plasma material interaction (PMI) conditions of the divertor of future magnetic confinement fusion power plants: energy flux 20MW/m 2 , ion fluence 1031/m 2 , pulse duration 106 sec. These goals of plasma exposure in MPEX are well beyond those achieved in magnetic fusion experimental devices. Successfully achieving these high power steady state conditions for long pulses requires operational control of the heating and particle sources and the plasma flux to the walls and target. The MPEX AI Hot Spot Controller, proposed in this project, will help achieve the operational milestones of MPEX. The MPEX device will begin commissioning at the end of FY26. A smaller proto-MPEX was operated for 14,666 plasma discharges and will resume operation in September of 2025 as proto-MPEX-lite, with reduced capability, to test a new window for the Helicon plasma source. The proto-MPEX data has undergone surrogate modeling with machine learning methods (R. Archibald, 2022 IEEE International Conference on Big Data). This proto-MPEX data will be used to begin development of the AI digital twins described in this white paper. The scientific mission of MPEX is to qualify materials of different composition for use in the high energy and plasma flux conditions of a fusion power plant. The materials exposed in MPEX will in some cases be exposed to high neutron fluxes at other ORNL facilities to measure the changes to their PMI properties. The targets exposed in MPEX will be transported under vacuum to a Surface Analysis Station (SAS). The SAS will be equipped with the following diagnostics: Focused Ion Beam (FIB) for trench milling, 100-400 angstrom resolution scanning electron microscope (SEM), surface mapping x-ray spectrometer, high resolution camera, and a future upgrade to a laser induced breakdown spectroscopy quadruple mass spectrometer (LIBS-QMS). The MPEX experiments will generate diverse pre- and post-exposure measurement data of detailed material properties down to the crystal grain level in 3D for post-exposure assessment of PMI damage (e.g. cracking, melting, erosion and redeposition of the material). Physics models for the PMI, and how the material composition and manufacturing impact its performance under high energy plasma exposure, need to be validated with MPEX data to guide the selection of new candidate materials. Our vision for the MPEX AI Digital Twins project is to supply experimental and physics model simulation data to train Artificial Intelligence (AI) models for data processing, analysis, operational control, PMI and materials simulation to maximize the scientific output of the MPEX device. Ultimately, an AI digital twin of MPEX material assessment metrics for tested and synthetic material types with simulated PMI will be trained by the AI Modeling Teams on the experimental and physics simulation data submitted to the American Science Cloud by this project. A purely empirical search for the best material is inefficient given the finite number of samples that can be tested on MPEX. In order to expand the material properties database for training the MPEX Material Assessment AI Digital Twin, and to gain physics understanding of the PMI processes, physics models of the material properties and PMI processes are required. The physics simulations provide detailed simulation data, like impact angles for plasma ions, sputtering yields, transport of the ionized sputtered target material in the plasma, and redeposition locations. This simulation data expands the measurement data for deeper physics understanding. The experimental data is essential to validate the PMI and material structure simulation models. The validated models can then be used to generate new simulation data of MPEX material assessments for synthetic material compositions that have not been exposed in MPEX. These predictive simulations, plus the whole experimental dataset, will be used to train the MPEX Material Assessment AI Digital Twin allowing a rapid generative AI search for new materials with reduced PMI damage by interpolating the domain of the training set. These new optimum materials can be simulated with the physics codes and/or tested in MPEX. The ability of AI neural networks to interpolate multi-dimensional parameter spaces and generate virtual data is exploited for a more efficient search for optimum materials. The advent of the Transformational AI Models Consortium (TAIMC) is an opportunity to engage with state of the art private and public AI developers to achieve the goals of the AI digital twins and AI accelerated physics models proposed in this project. Our partners at ORNL from the Advance Scientific Computing Research (ASCR) organization will collaborate in accelerating the integrated plasma material interaction simulation framework. This simulation framework will provide a platform for generating simulation data across a range of physical fidelities, including hybrid methods that produce multi-fidelity results. This data will be leveraged for AI model development, both for generation of surrogates and the automation of simulation campaigns. A part of the research below will include collaborative efforts with the TAIMC to (i) adapt data storage approaches to ensure AI-readiness, (ii) provide a protypical exemplar to inform and exercise constructed workflows, and (iii) generate and share data, using the TAIMC unified AI data standard, for foundational models that will be trained from multiple sources across the DOE complex. We will also collaborate with the TAIMC, as well as the planned AI modeling teams, to develop approaches for reducing the cost of data generation. These include tailored multi-fidelity approaches as well as fine-tuning strategies to augment general, large-scale foundational models.
The Virtual Federated Science Instrument Environment (VFSIE) is a digital twin of a federated infrastructure of multiple sites with science instruments and computing systems that are geographically dispersed. It emulates the federation using containers and virtual hosts (vhosts) connected over local and wide-area networks for the main purpose of supporting the development and testing of science workflows and software stacks, which will subsequently transition to the physical infrastructure. In this demo, we present VFSIE of a four sites scenario and illustrate the following use cases: (i) execution of federation stack components that expose resources to enable a science user to execute workflows, (ii) execution of remote control commands of beam-line instruments to collect measurements and position the sensor, and (iii) remote execution and access of Jupyter Notebook for tomographic image computations