Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “operator networks”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Stacked networks improve physics-informed training: Applications to neural networks and deep operator networks

Physics-informed neural networks and operator networks have shown promise for effectively solving equations modeling physical systems. However, these networks can happen to be difficult or impossible to train accurately. Here, we present a novel multifidelity framework for stacking physics-informed neural networks and operator networks that facilitates training. We successively build a chain of networks, where the output at one step can act as a low-fidelity input for training a longer chain, gradually increasing the expressivity of the learnt model. The equations imposed at each step of the iterative process can be the same or different (akin to simulated annealing). The iterative (stacking) nature of the proposed method allows us to learn progressively features of a solution which could have been hard to learn directly. Through benchmark problems including a nonlinear pendulum, the wave equation, and the viscous Burgers equation, we show how stacking can be used to improve the accuracy and reduce the required size of physics-informed neural networks and operator networks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Self-adaptive weights based on balanced residual decay rate for physics-informed neural networks and deep operator networks

Physics-informed deep learning has emerged as a promising alternative for solving partial differential equations. However, for complex problems, training these networks can still be challenging, often resulting in unsatisfactory accuracy and efficiency. In this work, we demonstrate that the failure of plain physics-informed neural networks arises from the significant discrepancy in the convergence rate of residuals at different training points, where the slowest convergence rate dominates the overall solution convergence. Based on these observations, we propose a pointwise adaptive weighting method that balances the residual decay rate across different training points. The performance of our proposed adaptive weighting method is compared with current state-of-the-art adaptive weighting methods on benchmark problems for both physics-informed neural networks and physics-informed deep operator networks. In conclusion, through extensive numerical results we demonstrate that our proposed approach of balanced residual decay rates offers several advantages, including bounded weights, high prediction accuracy, fast convergence rate, low training uncertainty, low computational cost, and ease of hyperparameter tuning.

Balanced convergence rate↗
NASA NTRS1990

L-band briefcase terminal network operation

During 1989, SkyWave Electronics Ltd. developed a lightweight battery powered, L Band Briefcase Satellite Terminal (LBT) which is capable of providing truly portable voice and data communications on a global scale. The LBT is designed to operate through Inmarsat's Atlantic Region Satellite MARECSB-2, and Teleglobe Canada's 18 meter Laurentides Earth Station. A small operating network, consisting of up to 20 mobile terminals and homing on the Laurentides Earth Station, was set up in the spring of 1990 to provide commercial service to LBT users, both domestic and international. This paper describes the LBT and the operation of the network.

Rossiter, P.↗
NASA NTRS2006

A Space Operations Network Alternative: Using Globally Connected Research and Education Networks for Space-Based Science Operations

Earth based networking in support of various space agency projects has been based on leased service/circuits which has a high associated cost. This cost is almost always taken from the science side resulting in less science. This is a proposal to use Research and Education Networks (RENs) worldwide to support space flight operations in general and space-based science operations in particular. The RENs were developed to support scientific and educational endeavors. They do not provide support for general Internet traffic. The connectivity and performance of the research and education networks is superb. The connectivity at Layer 3 (IP) virtually encompasses the globe. Most third world countries and all developed countries have their own research and education networks, which are connected globally. Performance of the RENs especially in the developed countries is exceptional. Bandwidth capacity currently exists and future expansion promises that this capacity will continue. REN performance statistics has always exceeded minimum requirements for spaceflight support. Research and Education networks are more loosely managed than a corporate network but are highly managed when compared to the commodity Internet. Management of RENs on an international level is accomplished by the International Network Operations Center at Indiana University at Indianapolis. With few exceptions, each regional and national REN has its own network ops center. The acceptable use policies (AUP), although differing by country, allows any scientific program or project the use of their networks. Once in compliance with the first RENs AUP, all others will accept that specific traffic including regional and transoceanic networks. RENs can support spaceflight related scientific programs and projects. Getting the science to the researcher is obviously key to any scientific project. RENs provide a pathway to virtually any college or university in the world, as well as many governmental institutes and science centers. RENs are not to be used for mission critical types of network traffic, even though RENs performance characteristics would support it.

Bradford, Robert N.↗
NASA NTRS1982

NASA Deep Space Network Operations Scheduling

The functioning of the Deep Space Network Operations Scheduling, Jet Propulsion Laboratory, CA is reviewed. The primary objectives of the Operations Scheduling are: to schedule the worldwide global allocation of ground communications, tracking facilities, and equipment; and to provide deep space telecommunications for command, tracking, telemetry, and control in support of flight mission operations and tests. Elements of the earth set are Deep Space Stations (DSS) which provide the telecommunications link between the earth and spacecraft; NASA Communications Network; Network Data Processing Area; Network Operations Control Area which provides operational direction to the DSS; Mission Control and Computing systems; and Mission Support areas which provide flight control of the spacecraft. Elements of the space set include mission priorities and requirements which determine the spacecraft queue for allocating network resources. Scheduling is discussed in terms of long-range (3 years), mid-range (8 weeks), and short-range (2 weeks).

Enari, D. M.↗
NASA NTRS1985

Network operating system

Long-term and short-term objectives for the development of a network operating system for the Space Station are stated. The short-term objective is to develop a prototype network operating system for a 100 megabit/second fiber optic data bus. The long-term objective is to establish guidelines for writing a detailed specification for a Space Station network operating system. Major milestones are noted. Information is given in outline form.

Source record↗
NASA NTRS1991

Unattended network operations technology assessment study. Technical support for defining advanced satellite systems concepts

The results are summarized of an unattended network operations technology assessment study for the Space Exploration Initiative (SEI). The scope of the work included: (1) identified possible enhancements due to the proposed Mars communications network; (2) identified network operations on Mars; (3) performed a technology assessment of possible supporting technologies based on current and future approaches to network operations; and (4) developed a plan for the testing and development of these technologies. The most important results obtained are as follows: (1) addition of a third Mars Relay Satellite (MRS) and MRS cross link capabilities will enhance the network's fault tolerance capabilities through improved connectivity; (2) network functions can be divided into the six basic ISO network functional groups; (3) distributed artificial intelligence technologies will augment more traditional network management technologies to form the technological infrastructure of a virtually unattended network; and (4) a great effort is required to bring the current network technology levels for manned space communications up to the level needed for an automated fault tolerance Mars communications network.

Price, Kent M.↗
NASA NTRS1982

Operations of the reconfigured Ground Communications Facility Central Communications Terminal and the Network Operations Control Center

The ground communications facility (GCF) central communications terminal and network operations control center (NOCC) hardware was rearranged, supplemented, and modified, and software programs changed to provide an improved GCF and NOCC operational environment and capability. Control center operations section activities required to make the changeover from the old to the new GCF and NOCC configuration are addressed.

Santana, J. C.↗
NASA NTRS1991

Technologies for unattended network operations

The necessary network management functions for a telecommunications, navigation and information management (TNIM) system in the framework of an extension of the ISO model for communications network management are described. Various technologies that could substantially reduce the need for TNIM network management, automate manpower intensive functions, and deal with synchronization and control at interplanetary distances are presented. Specific technologies addressed include the use of the ISO Common Management Interface Protocol, distributed artificial intelligence for network synchronization and fault management, and fault-tolerant systems engineering.

Jaworski, Allan↗
DOE OSTI2025

Efficient Training of Deep Neural Operator Networks via Randomized Sampling

Neural operators (NOs) employ deep neural networks to learn the mappings between infinitedimensional function spaces. Deep operator network (DeepONet), a popular NO architecture, has demonstrated success in the real-time prediction of complex dynamics across various scientific and engineering applications. In this work, we introduce a random sampling technique to be adopted during the training of DeepONet, aimed at improving the generalization ability of the model, while significantly reducing the computational time. The proposed approach targets the trunk network of the DeepONet model that outputs the basis functions corresponding to the spatiotemporal locations of the bounded domain on which the physical system is defined. While constructing the loss function, DeepONet training traditionally considers a uniform grid of spatiotemporal points at which all the output functions are evaluated for each iteration. This approach leads to a larger batch size, resulting in poor generalization and increased memory demands, due to the limitations of the stochastic gradient descent (SGD) optimizer. The proposed random sampling over the inputs of the trunk net mitigates these challenges, improving generalization and reducing the memory requirements during training, resulting in significant computational gains. We validate our hypothesis through three benchmark examples, demonstrating substantial reductions in training time while achieving comparable or lower overall test errors relative to the traditional training approach. Our results indicate that incorporating randomization in the trunk network inputs during training enhances the efficiency and robustness of DeepONet, offering a promising avenue for improving the framework’s performance in modeling complex physical systems.

Karumuri, Sharmila [Department of Civil & Systems ↗
DOE OSTI2026

Neural Scaling Laws of Deep ReLU and Deep Operator Network: A Theoretical Study

Neural scaling laws play a pivotal role in the performance of deep neural networks and have been observed in a wide range of tasks. However, a complete theoretical framework for understanding these scaling laws remains underdeveloped. In this paper, we explore the neural scaling laws for deep operator networks, which involve learning mappings between function spaces, with a focus on the Chen and Chen style architecture. These approaches, which include the popular Deep Operator Network (DeepONet), approximate the output functions using a linear combination of learnable basis functions and coefficients that depend on the input functions. We establish a theoretical framework to quantify the neural scaling laws by analyzing its approximation and generalization errors. We articulate the relationship between the approximation and generalization errors of deep operator networks and key factors such as network model size and training data size. Moreover, we address cases where input functions exhibit low-dimensional structures, allowing us to derive tighter error bounds. These results also hold for deep ReLU networks and other similar structures. Our results offer a partial explanation of the neural scaling laws in operator learning and provide a theoretical foundation for their applications.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1978

Network Operations Control Center Block 3 modifications

The Network Operations Control Center (NOCC) Block 3 configuration is in the process of being upgraded to provide the capabilities required to support Voyager and Pioneer Venus 1978 Project commitments as well as to support Deep Space Station changes arising from the Mark 3 DSN Data Subsystem implementation. Information is given on the hardware and software changes necessary to implement such capabilities.

Garcia, E. A.↗
NASA NTRS2010

Using Web 2.0 Techniques in NASA's Ares Engineering Operations Network (AEON) Environment - First Impressions

The Mission Operations Laboratory (MOL) at Marshall Space Flight Center (MSFC) is responsible for Engineering Support capability for NASA s Ares rocket development and operations. In pursuit of this, MOL is building the Ares Engineering and Operations Network (AEON), a web-based portal to support and simplify two critical activities: Access and analyze Ares manufacturing, test, and flight performance data, with access to Shuttle data for comparison Establish and maintain collaborative communities within the Ares teams/subteams and with other projects, e.g., Space Shuttle, International Space Station (ISS). AEON seeks to provide a seamless interface to a) locally developed engineering applications and b) a Commercial-Off-The-Shelf (COTS) collaborative environment that includes Web 2.0 capabilities, e.g., blogging, wikis, and social networking. This paper discusses how Web 2.0 might be applied to the typically conservative engineering support arena, based on feedback from Integration, Verification, and Validation (IV&V) testing and on searching for their use in similar environments.

Scott, David W.↗
NASA NTRS1993

Deep Space Network (DSN), Network Operations Control Center (NOCC) computer-human interfaces

The Network Operations Control Center (NOCC) of the DSN is responsible for scheduling the resources of DSN, and monitoring all multi-mission spacecraft tracking activities in real-time. Operations performs this job with computer systems at JPL connected to over 100 computers at Goldstone, Australia and Spain. The old computer system became obsolete, and the first version of the new system was installed in 1991. Significant improvements for the computer-human interfaces became the dominant theme for the replacement project. Major issues required innovating problem solving. Among these issues were: How to present several thousand data elements on displays without overloading the operator? What is the best graphical representation of DSN end-to-end data flow? How to operate the system without memorizing mnemonics of hundreds of operator directives? Which computing environment will meet the competing performance requirements? This paper presents the technical challenges, engineering solutions, and results of the NOCC computer-human interface design.

Ellman, Alvin↗
NASA NTRS1993

GPS Global Tracking Network Operations At JPL During the 1992 IGS Experiment and Beyond

The GPS Network Operations center at JPL Provides resources for the collection, distribution and maintenance of GPS data from global network of GPS receivers. This includes the ability to remotely diagnosis receiver problems, communicate with receivers and operators of receivers, to automatically collect, process and distribute data for 14 stations in the IGS network. This amounts to 35 megabytes of raw, finished data and products daily.

GPS receiver Global Network↗
NASA NTRS1985

State of the art survey of network operating systems development

The results of the State-of-the-Art Survey of Network Operating Systems (NOS) performed for Goddard Space Flight Center are presented. NOS functional characteristics are presented in terms of user communication data migration, job migration, network control, and common functional categories. Products (current or future) as well as research and prototyping efforts are summarized. The NOS products which are revelant to the space station and its activities are evaluated.

Source record↗