Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed architecture”

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 829 records · Page 46

Enhanced Control, Optimization, and Integration of Distributed Energy Applications (ECO-IDEA)

With support from the U.S. Department of Energy Solar Energy Technologies Office, the National Renewable Energy Laboratory (NREL) partnered with Xcel Energy, Schneider Electric, Varentec, and Electric Power Research Institute (EPRI) to meet the goals of the Enabling Extreme Real-Time Grid Integration of Solar Energy (ENERGISE) program. This project developed and validated an innovative data-enhanced hierarchical control architecture that enables the efficient, reliable, resilient, and secure operation of future distribution systems with a high penetration of distributed energy resources like solar energy. The architecture enables a hybrid control approach where a centralized control layer is complemented by distributed control algorithms for solar inverters and autonomous control of grid edge devices. It is fully interoperable and includes all the cybersecurity aspects necessary for reliable and secure system operation. The hybrid approach can seamlessly integrate multiple voltage-regulation technologies, both at central and grid-edge levels, which enables reliable and efficient system operation in the face of unpredictable conditions. The overarching goal of the Eco-Idea project is to develop, validate, and deploy a unique and innovative Data-Enhanced Hierarchical Control (DEHC) architecture that comprehensively addresses the formidable challenges associated with proliferation of high penetration of distributed PV such as reverse power flows, transients from variability of PV systems, feeder load balancing, and voltage stability. These issues are exposing the weaknesses of existing grid operations and controls - including, but not limited to, lack of grid situational awareness, heuristic and slow-acting control actions, latency of control for emergency situations, and points of failure in communications. The proposed architecture will comprehensively resolve the deficiencies of current operational settings - where monitoring and control solutions proposed across industry and academia may not be interoperable and may not coexist in the same system - and will enable an efficient, reliable, resilient, and secure operation of future distribution systems with penetration of solar energy well beyond current limits. The DEHC architecture was developed and validated rigorously through hardware-in-loop simulations in the laboratory environment and deployed on the field.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Virtual Power Plant Architecture and Resilient Design

Virtual Power Plants (VPPs) represent a fundamental shift in electric grid operations, aggregating distributed energy resources (DERs) such as solar panels and battery storage to deliver utility-scale grid services traditionally provided by centralized power plants. This report examines the unique architectural, operational, and digital assurance considerations that distinguish VPPs from conventional utility infrastructure as they scale from pilot projects to mainstream deployment across the United States. While VPPs offer significant opportunities for grid modernization and enhanced flexibility, their distributed, multi-stakeholder architecture introduces distinct security challenges that differ fundamentally from traditional generation facilities. The analysis identifies risks in VPP operations, including device-level security gaps, platform vulnerabilities, and communication protocol weaknesses that create expanded attack surfaces compared to centralized power plants. Through examination of real-world incidents and emerging threat patterns, the report demonstrates how some VPPs' reliance on consumer-owned devices, public internet infrastructure, and complex vendor ecosystems require new approaches to digital assurance and operational security. The findings provide practical guidance for utilities, regulators, and aggregators to implement robust security frameworks and operational best practices essential for maintaining grid reliability as VPP deployment accelerates under the Federal Energy Regulatory Commission (FERC) Order 2222 and related regulatory initiatives.

24 - POWER TRANSMISSION AND DISTRIBUTION↗

Explainable Neural Architecture Search (XNAS)

Code for the paper Learning Interpretable Models Through Multi-Objective Neural Architecture Search by Zachariah Carmichael, Tim Moon, and Sam Ade Jacobs. Monumental advances in deep learning have led to unprecedented achievements across a multitude of domains. While the performance of deep neural networks is indubitable, the architectural design and interpretability of such models are nontrivial. Research has been introduced to automate the design of neural network architectures through neural architecture search (NAS). Recent progress has made these methods more pragmatic by exploiting distributed computation and novel optimization algorithms. However, there is little work in optimizing architectures for interpretability. To this end, we propose a multiobjective distributed NAS framework that optimizes for both task performance and introspection. We leverage the non-dominated sorting genetic algorithm (NSGA-II) and explainable AI (XAI) techniques to reward architectures that can be better comprehended by humans. The framework is evaluated on several image classification datasets. We demonstrate that jointly optimizing for introspection ability and task error leads to more disentangled architectures that perform within tolerable error.

Carmichael, ZachariahJ↗

Discordant nature of Cd in PbSe: off-centering and core–shell nanoscale CdSe precipitates lead to high thermoelectric performance

We report a novel hierarchical microstructure in the PbSe–CdSe system, which collectively contributes to significant enhancement in thermoelectric performance, with ZT ave ~ 0.83 across the 400–923 K temperature range, the highest reported for p-type, Te-free PbSe systems. We have investigated the local atomic structure as well as the microstructure of a series of PbSe–xCdSe materials, up to x = 10%. We find that the behavior of the Cd atoms in the octahedral rock salt sites is discordant and results in off-center displacement and distortion. Such off-centered Cd in the PbSe matrix creates (1) L–Σ electronic energy band convergence, (2) a flattened L band, both contributing to higher Seebeck coefficients, and (3) enhanced phonon scattering, which leads to lower thermal conductivity. These conclusions are supported by photoemission yield spectroscopy in air (PYSA), solid state 111 Cd, 77 Se NMR spectroscopy and DFT calculations. Above the solubility limit (>6%CdSe), we also observe endotaxial CdSe nano-precipitates with core–shell architecture formed in PbSe, whose size, distribution and structure gradually change with the Cd content. The nano-precipitates exhibit a zinc blende crystal structure and a tetrahedral shape with significant local strain, but are covered with a thin wurtzite layer along the precipitate/matrix interface, creating a core–shell structure embedded in PbSe. This newly discovered architecture causes a further reduction in lattice thermal conductivity. Furthermore, potassium is found to be an effective p-type dopant in the PbSe–CdSe system, leading to an enhanced power factor, a maximum ZT of ~1.4 at 923 K for Pb 0.98 K 0.02 Se–6%CdSe.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Sampling Functions from Gaussian Processes and Structured Covariance Gaussian Networks

When learning aerodynamic models from data, it is critical to incorporate estimates of model uncertainty. This motivates the design of probabilistic aerodynamic databases which can be sampled to generate physically and statistically plausible aerodynamic models. In this talk we discuss how to sample deterministic functions from two different kinds of probabilistic models and demonstrate their use. First, Gaussian Process Regressors (GPRs) are a widely used probabilistic kernel-based model which can be thought of as Gaussian distributions over functions. GPRs are generally trained by maximizing the marginal likelihood of seeing the training data over the kernel parameter space. Sample functions are easily generated by drawing points from the Gaussian distribution at desired input points. However, when the points are not known ahead of time, the classical sampling approach is not possible since successive function samples will generate different function realizations. We present an approach for sampling consistent function evaluations from a GPR over multiple samples. Second, we describe a neural network architecture which learns a conditional Gaussian distribution by maximizing the marginal likelihood at each point in the input space. We then discuss and compare several options for generating sample functions which match this distribution. Finally, we demonstrate the use of these probabilistic aerodynamic models in an atmospheric reentry simulation.

Gaussian process regression↗

Fiber optic computing using distributed feedback

Abstract The widespread adoption of machine learning and other matrix intensive computing algorithms has renewed interest in analog optical computing, which has the potential to perform large-scale matrix multiplications with superior energy scaling and lower latency than digital electronics. However, most optical techniques rely on spatial multiplexing, requiring a large number of modulators and detectors, and are typically restricted to performing a single kernel convolution operation per layer. Here, we introduce a fiber-optic computing architecture based on temporal multiplexing and distributed feedback that performs multiple convolutions on the input data in a single layer. Using Rayleigh backscattering in standard single mode fiber, we show that this technique can efficiently apply a series of random nonlinear projections to the input data, facilitating a variety of computing tasks. The approach enables efficient energy scaling with orders of magnitude lower power consumption than GPUs, while maintaining low latency and high data-throughput.

97 MATHEMATICS AND COMPUTING↗

A Grid Infrastructure for Supporting Space-based Science Operations

Emerging technologies for computational grid infrastructures have the potential for revolutionizing the way computers are used in all aspects of our lives. Computational grids are currently being implemented to provide a large-scale, dynamic, and secure research and engineering environments based on standards and next-generation reusable software, enabling greater science and engineering productivity through shared resources and distributed computing for less cost than traditional architectures. Combined with the emerging technologies of high-performance networks, grids provide researchers, scientists and engineers the first real opportunity for an effective distributed collaborative environment with access to resources such as computational and storage systems, instruments, and software tools and services for the most computationally challenging applications.

Bradford, Robert N.↗

Using Ada for a distributed, fault tolerant system

It is pointed out that advanced avionics applications increasingly require underlying machine architectures which are damage and fault tolerant, and which provide access to distributed sensors, effectors and high-throughput computational resources. The Advanced Information Processing System (AIPS), sponsored by NASA, is to provide an architecture which can meet the considered requirements. Ada was selected for implementing the AIPS system software. Advantages of Ada are related to its provisions for real-time programming, error detection, modularity and separate compilation, and standardization and portability. Chief drawbacks of this language are currently limited availability and maturity of language implementations, and limited experience in applying the language to real-time applications. The present investigation is concerned with current plans for employing Ada in the design of the software for AIPS. Attention is given to an overview of AIPS, AIPS software services, and representative design issues in each of four major software categories.

Dewolf, J. B.↗

A Distributed Simulation Facility to Support Human Factors Research in Advanced Air Transportation Technology

A distributed real-time simulation of the civil air traffic environment developed to support human factors research in advanced air transportation technology is presented. The distributed environment is based on a custom simulation architecture designed for simplicity and flexibility in human experiments. Standard Internet protocols are used to create the distributed environment, linking all advanced cockpit simulator, all Air Traffic Control simulator, and a pseudo-aircraft control and simulation management station. The pseudo-aircraft control station also functions as a scenario design tool for coordinating human factors experiments. This station incorporates a pseudo-pilot interface designed to reduce workload for human operators piloting multiple aircraft simultaneously in real time. The application of this distributed simulation facility to support a study of the effect of shared information (via air-ground datalink) on pilot/controller shared situation awareness and re-route negotiation is also presented.

Amonlirdviman, Keith↗

Cathode architectures for alkali metal / oxygen batteries

Electrochemical energy storage devices, such as alkali metal-oxygen battery cells (e.g., non-aqueous lithium-air cells), have a cathode architecture with a porous structure and pore composition that is tailored to improve cell performance, especially as it pertains to one or more of the discharge/charge rate, cycle life, and delivered ampere-hour capacity. A porous cathode architecture having a pore volume that is derived from pores of varying radii wherein the pore size distribution is tailored as a function of the architecture thickness is one way to achieve one or more of the aforementioned cell performance improvements.

25 ENERGY STORAGE↗

Avionics architecture for the U.S. segment of the international space station alpha

The International Space Station Alpha (ISSA) is a joint project between the United States, European Space Agency, Japan, Canada and Russia to develop and fly a space station in the later part of the 1990s. The United States will be the largest hardware contributor to this project, and this paper will describe the avionics architecture for the U.S. provided pressurized modules and truss segments. The avionics contained within these U.S. modules and segments will perform many of the core engineering and payload support functions for the ISSA. These functions include: the command and data handling, communications and tracking, guidance navigation and control (in partnership with the Russian elements), and electrical power generation and distribution. This paper will describe the architecture of the avionics systems that will perform these core engineering avionics functions, as well as provide a discussion on the avionics necessary for the proper functioning of the core engineering avionics supported functions such as thermal control, life support, and mechanisms.

Smith, Joseph F.↗

Combined Impact of Heterogeneous Lifetime and Gettering on Solar Cell Performance

We couple numerical process and device simulations to provide a framework for understanding the combined effects of as-grown wafer impurity distribution, processing parameters, and solar cell architecture. For this study, we added the Impurity-to-Efficiency simulator to Synopsys’ Sentaurus Process software using the Alagator Scripting Language. Our results quantify how advanced processing can eliminate differences in efficiency due to different as-grown impurity concentrations and due to different area fractions of defective wafer regions. We identify combinations of as-grown impurity distributions and process parameters that produce solar cells limited by point defects and those that are limited by precipitated impurities. Gettering targeted at either point defect or precipitate reduction can then be designed and applied to increase cell efficiency. We also visualize the post-processing iron and total recombination distributions in 2D maps of the wafer cross-section. PV researchers and companies can input their initial iron distributions and processing parameters into our software and couple the resulting process simulation results with a solar cell device design of interest to conduct their own analyses. The Alagator scripts we developed are freely available online at http://pv.mit.edu/impurity-to-efficiency-i2e-simulator-for-sentaurus-tcad/.

14 SOLAR ENERGY↗

Taking advantage of ground data systems attributes to achieve quality results in testing software

During the software development life cycle process, basic testing starts with the development team. At the end of the development process, an acceptance test is performed for the user to ensure that the deliverable is acceptable. Ideally, the delivery is an operational product with zero defects. However, the goal of zero defects is normally not achieved but is successful to various degrees. With the emphasis on building low cost ground support systems while maintaining a quality product, a key element in the test process is simulator capability. This paper reviews the Transportable Payload Operations Control Center (TPOCC) Advanced Spacecraft Simulator (TASS) test tool that is used in the acceptance test process for unmanned satellite operations control centers. The TASS is designed to support the development, test and operational environments of the Goddard Space Flight Center (GSFC) operations control centers. The TASS uses the same basic architecture as the operations control center. This architecture is characterized by its use of distributed processing, industry standards, commercial off-the-shelf (COTS) hardware and software components, and reusable software. The TASS uses much of the same TPOCC architecture and reusable software that the operations control center developer uses. The TASS also makes use of reusable simulator software in the mission specific versions of the TASS. Very little new software needs to be developed, mainly mission specific telemetry communication and command processing software. By taking advantage of the ground data system attributes, successful software reuse for operational systems provides the opportunity to extend the reuse concept into the test area. Consistency in test approach is a major step in achieving quality results.

Sigman, Clayton B.↗

High Performance FORTRAN

High performance FORTRAN is a set of extensions for FORTRAN 90 designed to allow specification of data parallel algorithms. The programmer annotates the program with distribution directives to specify the desired layout of data. The underlying programming model provides a global name space and a single thread of control. Explicitly parallel constructs allow the expression of fairly controlled forms of parallelism in particular data parallelism. Thus the code is specified in a high level portable manner with no explicit tasking or communication statements. The goal is to allow architecture specific compilers to generate efficient code for a wide variety of architectures including SIMD, MIMD shared and distributed memory machines.

Mehrotra, Piyush↗

Distributed Data-Flow for In-Situ Visualization and Analysis at Petascale

We conducted a feasibility study to research modifications to data-flow architectures to enable data-flow to be distributed across multiple machines automatically. Distributed data-flow is a crucial technology to ensure that tools like the VisIt visualization application can provide in-situ data analysis and post-processing for simulations on peta-scale machines. We modified a version of VisIt to study load-balancing trade-offs between light-weight kernel compute environments and dedicated post-processing cluster nodes. Our research focused on memory overheads for contouring operations, which involves variable amounts of generated geometry on each node and computation of normal vectors for all generated vertices. Each compute node independently decided whether to send data to dedicated post-processing nodes at each stage of pipeline execution, depending on available memory. We instrumented the code to allow user settable available memory amounts to test extremely low-overhead compute environments. We performed initial testing of this prototype distributed streaming framework, but did not have time to perform scaling studies at and beyond 1000 compute-nodes.

97 MATHEMATICS AND COMPUTING↗

An Architecture for Real-Time Interpretation and Visualization of Structural Sensor Data in a Laboratory Environment

A visualization system is being developed out of the need to monitor, interpret, and make decisions based on the information from several thousand sensors during experimental testing to facilitate development and validation of structural health monitoring algorithms. As an added benefit the system will enable complete real-time sensor assessment of complex test specimens. Complex structural specimens are routinely tested that have hundreds or thousands of sensors. During a test, it is impossible for a single researcher to effectively monitor all the sensors and subsequently interesting phenomena occur that are not recognized until post-test analysis. The ability to detect and alert the researcher to these unexpected phenomena as the test progresses will significantly enhance the understanding and utilization of complex test articles. Utilization is increased by the ability to halt a test when the health monitoring algorithm response is not satisfactory or when an unexpected phenomenon occurs, enabling focused investigation potentially through the installation of additional sensors. Often if the test continues, structural changes make it impossible to reproduce the conditions that exhibited the phenomena. The prohibitive time and costs associated with fabrication, sensoring, and subsequent testing of additional test articles generally makes it impossible to further investigate the phenomena. A scalable architecture is described to address the complex computational demands of structural health monitoring algorithm development and laboratory experimental test monitoring. The researcher monitors the test using a photographic quality 3D graphical model with actual sensor locations identified. In addition, researchers can quickly activate plots displaying time or load versus selected sensor response along with the expected values and predefined limits. The architecture has several key features. First, distributed dissimilar computers may be seamlessly integrated into the information flow. Second, virtual sensors may be defined that are complex functions of existing sensors or other virtual sensors. Virtual sensors represent a calculated value not directly measured by particular physical instrument. They can be used, for example, to represent the maximum difference in a range of sensors or the calculated buckling load based on the current strains. Third, the architecture enables autonomous response to preconceived events, where by the system can be configured to suspend or abort a test if a failure is detected in the load introduction system. Fourth, the architecture is designed to allow cooperative monitoring and control of the test progression from multiple stations both remote and local to the test system. To illustrate the architecture, a preliminary implementation is described monitoring the Stitched Composite Wing recently tested at LaRC.

Doggett, William↗

Performance Assessment of OVERFLOW on Distributed Computing Environment

The aerodynamic computer code, OVERFLOW, with a multi-zone overset grid feature, has been parallelized to enhance its performance on distributed and shared memory paradigms. Practical application benchmarks have been set to assess the efficiency of code's parallelism on high-performance architectures. The code's performance has also been experimented with in the context of the distributed computing paradigm on distant computer resources using the Information Power Grid (IPG) toolkit, Globus. Two parallel versions of the code, namely OVERFLOW-MPI and -MLP, have developed around the natural coarse grained parallelism inherent in a multi-zonal domain decomposition paradigm. The algorithm invokes a strategy that forms a number of groups, each consisting of a zone, a cluster of zones and/or a partition of a large zone. Each group can be thought of as a process with one or multithreads assigned to it and that all groups run in parallel. The -MPI version of the code uses explicit message-passing based on the standard MPI library for sending and receiving interzonal boundary data across processors. The -MLP version employs no message-passing paradigm; the boundary data is transferred through the shared memory. The -MPI code is suited for both distributed and shared memory architectures, while the -MLP code can only be used on shared memory platforms. The IPG applications are implemented by the -MPI code using the Globus toolkit. While a computational task is distributed across multiple computer resources, the parallelism can be explored on each resource alone. Performance studies are achieved with some practical aerodynamic problems with complex geometries, consisting of 2.5 up to 33 million grid points and a large number of zonal blocks. The computations were executed primarily on SGI Origin 2000 multiprocessors and on the Cray T3E. OVERFLOW's IPG applications are carried out on NASA homogeneous metacomputing machines located at three sites, Ames, Langley and Glenn. Plans for the future will exploit the distributed parallel computing capability on various homogeneous and heterogeneous resources and large scale benchmarks. Alternative IPG toolkits will be used along with sophisticated zonal grouping strategies to minimize the communication time across the computer resources.

Djomehri, M. Jahed↗

Compressed basis GMRES on high-performance graphics processing units

Krylov methods provide a fast and highly parallel numerical tool for the iterative solution of many large-scale sparse linear systems. To a large extent, the performance of practical realizations of these methods is constrained by the communication bandwidth in current computer architectures, motivating the investigation of sophisticated techniques to avoid, reduce, and/or hide the message-passing costs (in distributed platforms) and the memory accesses (in all architectures). This article leverages Ginkgo’s memory accessor in order to integrate a communication-reduction strategy into the (Krylov) GMRES solver that decouples the storage format (i.e., the data representation in memory) of the orthogonal basis from the arithmetic precision that is employed during the operations with that basis. Given that the execution time of the GMRES solver is largely determined by the memory accesses, the cost of the datatype transforms can be mostly hidden, resulting in the acceleration of the iterative step via a decrease in the volume of bits being retrieved from memory. Together with the special properties of the orthonormal basis (whose elements are all bounded by 1), this paves the road toward the aggressive customization of the storage format, which includes some floating-point as well as fixed-point formats with mild impact on the convergence of the iterative process. We develop a high-performance implementation of the “compressed basis GMRES” solver in the Ginkgo sparse linear algebra library using a large set of test problems from the SuiteSparse Matrix Collection. We demonstrate robustness and performance advantages on a modern NVIDIA V100 graphics processing unit (GPU) of up to 50% over the standard GMRES solver that stores all data in IEEE double-precision.

97 MATHEMATICS AND COMPUTING↗