SEARCH · Engineering Papers
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.
Embodiments of the present disclosure are related to training one or more of machine learning algorithms in a virtual reality environment for error detection and correction and/or for employing one or more trained machine learning models in an augmented reality environment to detect and/or correct user errors associated the performance of one or more tasks.
Abstract Longitudinal phase space (LPS) provides a critical information about electron beam dynamics for various scientific applications. For example, it can give insight into the high-brightness X-ray radiation from a free electron laser. Existing diagnostics are invasive, and often times cannot operate at the required resolution. In this work we present a machine learning-based Virtual Diagnostic (VD) tool to accurately predict the LPS for every shot using spectral information collected non-destructively from the radiation of relativistic electron beam. We demonstrate the tool’s accuracy for three different case studies with experimental or simulated data. For each case, we introduce a method to increase the confidence in the VD tool. We anticipate that spectral VD would improve the setup and understanding of experimental configurations at DOE’s user facilities as well as data sorting and analysis. The spectral VD can provide confident knowledge of the longitudinal bunch properties at the next generation of high-repetition rate linear accelerators while reducing the load on data storage, readout and streaming requirements.
Abstract Real-time monitoring is a foundation of nuclear digital twin technology, crucial for detecting material degradation and maintaining nuclear system integrity. Traditional physical sensor systems face limitations, particularly in measuring critical parameters in hard-to-reach or harsh environments, often resulting in incomplete data coverage. Machine learning-driven virtual sensors offer a transformative solution by complementing physical sensors in monitoring critical degradation indicators. This paper introduces the use of Deep Operator Networks (DeepONet) to predict key thermal-hydraulic parameters in the hot leg of pressurized water reactor. DeepONet acts as a virtual sensor, mapping operational inputs to spatially distributed system behaviors without requiring frequent retraining. Our results show that DeepONet achieves low mean squared and Relative L2 error, making predictions 1400 times faster than traditional CFD simulations . These characteristics enable DeepONet to function as a real-time virtual sensor, synchronizing with the physical system to track degradation conditions and provide insights within the digital twin framework for nuclear systems.
The technology to reliably transmit high-resolution visual imagery over short to medium distances in real time has led to the serious considerations of the use of telemedicine, telepresence, and telerobotics in the delivery of health care. These concepts may involve, and evolve toward: consultation from remote expert teaching centers; diagnosis; triage; real-time remote advice to the surgeon; and real-time remote surgical instrument manipulation (telerobotics with virtual reality). Further extrapolation leads to teledesign and telereplication of spare surgical parts through quantitative teleimaging of 3-D surfaces tied to CAD/CAM devices and an artificially intelligent archival data base of 'normal' shapes. The ability to generate 'topogrames' or 3-D surface numerical tables of coordinate values capable of creating computer-generated virtual holographic-like displays, machine part replication, and statistical diagnostic shape assessment is critical to the progression of telemedicine. Any virtual reality simulation will remain in 'video-game' realm until realistic dimensional and spatial relational inputs from real measurements in vivo during surgeries are added to an ever-growing statistical data archive. The challenges of managing and interpreting this 3-D data base, which would include radiographic and surface quantitative data, are considerable. As technology drives toward dynamic and continuous 3-D surface measurements, presenting millions of X, Y, Z data points per second of flexing, stretching, moving human organs, the knowledge base and interpretive capabilities of 'brilliant robots' to work as a surgeon's tireless assistants becomes imaginable. The brilliant robot would 'see' what the surgeon sees--and more, for the robot could quantify its 3-D sensing and would 'see' in a wider spectral range than humans, and could zoom its 'eyes' from the macro world to long-distance microscopy. Unerring robot hands could rapidly perform machine-aided suturing with precision micro-sewing machines, splice neural connections with laser welds, micro-bore through constricted vessels, and computer combine ultrasound, microradiography, and 3-D mini-borescopes to quickly assess and trace vascular problems in situ. The spatial relationships between organs, robotic arms, and end-effector diagnostic, manipulative, and surgical instruments would be constantly monitored by the robot 'brain' using inputs from its multiple 3-D quantitative 'eyes' remote sensing, as well as by contact and proximity force measuring devices. Methods to create accurate and quantitative 3-D topograms at continuous video data rates are described.
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.
Neural networks are a very useful tool for analyzing and modeling complex real world systems. Applying neural network simulations to real world problems generally involves large amounts of data and massive amounts of computation. To efficiently handle the computational requirements of large problems, we have implemented at Los Alamos a highly efficient neural network compiler for serial computers, vector computers, vector parallel computers, and fine grain SIMD computers such as the CM-2 connection machine. This paper describes the mapping used by the compiler to implement feed-forward backpropagation neural networks for a SIMD (Single Instruction Multiple Data) architecture parallel computer. Thinking Machines Corporation has benchmarked our code at 1.3 billion interconnects per second (approximately 3 gigaflops) on a 64,000 processor CM-2 connection machine (Singer 1990). This mapping is applicable to other SIMD computers and can be implemented on MIMD computers such as the CM-5 connection machine. Our mapping has virtually no communications overhead with the exception of the communications required for a global summation across the processors (which has a sub-linear runtime growth on the order of O(log(number of processors)). We can efficiently model very large neural networks which have many neurons and interconnects and our mapping can extend to arbitrarily large networks (within memory limitations) by merging the memory space of separate processors with fast adjacent processor interprocessor communications. This paper will consider the simulation of only feed forward neural network although this method is extendable to recurrent networks.
Explore the source record for details and available documents.
As part of upgrading the Contact Dynamics Simulation Laboratory (CDSL) at the NASA Marshall Space Flight Center (MSFC), a simple, cost effective method was needed to communicate data among the networked simulation machines and I/O controllers used to run the facility. To fill this need and similar applicable situations, a generic protocol was developed, called ShareSync. ShareSync is a lightweight, real-time, publish-subscribe Ethernet protocol for simple and deterministic data sharing across diverse machines and operating systems. ShareSync provides a simple Application Programming Interface (API) for simulation programmers to incorporate into their code. The protocol is compatible with virtually all Ethernet-capable machines, is flexible enough to support a variety of applications, is fast enough to provide soft real-time determinism, and is a low-cost resource for distributed simulation development, deployment, and maintenance. The first design cycle iteration of ShareSync has been completed, and the protocol has undergone several testing procedures including endurance and benchmarking tests and approaches the 2001ts data synchronization design goal for the CDSL.
The development of an alternative medium for displaying information in complex human-machine interfaces is described. The 3-D virtual acoustic display is a means for accurately transferring information to a human operator using the auditory modality; it combines directional and semantic characteristics to form naturalistic representations of dynamic objects and events in remotely sensed or simulated environments. Although the technology can stand alone, it is envisioned as a component of a larger multisensory environment and will no doubt find its greatest utility in that context. The general philosophy in the design of the display has been that the development of advanced computer interfaces should be driven first by an understanding of human perceptual requirements, and later by technological capabilities or constraints. In expanding on this view, current and potential uses are addressed of virtual acoustic displays, such displays are characterized, and recent approaches to their implementation and application are reviewed, the research project at NASA-Ames is described in detail, and finally some critical research issues for the future are outlined.
Nuclear nonproliferation serves as a key goal, being undertaken by the International Atomic Energy Agency (IAEA). To recognize proliferation there are two pathways that states, who intend to use nuclear material for malicious purposes can take, diversion can misuse. Diversion is when fissile nuclear material is declared to the IAEA for non-weapon purposes, but then covertly removed. If the source of nuclear material, that is not declared and not fissionable, is placed inside the reactor core to create fissile material used to create weapons then the state is using the second pathway of proliferation, misuse. With the emerging development in areas of simulation and machine learning the creation of virtual models of reactor systems, digital twins, serve as a potential method to identify proliferation through detecting anomalous behavior in the reactor. A digital twin for a physical nuclear reactor has never been developed, as digital twins serve as an emerging technology. To investigate the process for development and use of a digital twin for a nuclear reactor Idaho State University’s AGN-201m serves as the nuclear reactor used for development of this digital twin. A data acquisition system has been installed to the reactor system allowing for the transfer of collected data from a reactor operation to Idaho National Laboratory’s Deeplynx data warehouse. When utilizing data to train reactor physics and machine learning models, a significant challenge encountered is the initial state of the data. Nuclear proliferation will have the capacity to be detected when the reactor immediately starts up, nor will it occur after the reactor shuts down. Generally, it will be detected when the reactor is operating at some desired power over a sufficient period for that specific reactor design. For the AGN-201m this will be when the reactor is critical (generally 1 mW or above) for a timespan that is within or less than the range of a regular business day. Datasets sent to Deeplynx have had to be manually cut to when the reactor is critical based on plots of power levels. This method is inefficient and laborious, especially when using multiple datasets at once to train a model. To provide a more streamlined approach an automated critical data extrapolator is developed, with capabilities of recognizing when the reactor operation first reaches criticality, and when the reactor undergoes a SCRAM and is shutdown.
This report presents a simulation based approach for sizing the axial balance elements of a pressure compensated external gear machine. The reference hydraulic unit must be able to guarantee operation in two quadrants, as either a pump or a motor. It also needs to operate in a wide operating range of speed and pressure, as it is to be used as primary hydraulic unit for an electro-hydraulic actuator (EHA). The design procedure builds upon past work at the author’s center, but it extends it to the case of multiple quadrant units. Also, a unique method for modeling of both mixed lubrication and cavitation that might occur in the lubricating film is used. After describing the design procedure, the paper discusses the features of the balancing of a reference unit, along with the effects of both cavitation and mixed lubrication.
We report that the age of cognitive computing and artificial intelligence (AI) is just dawning. Inspired by its successes and promises, several AI ecosystems are blossoming, many of them within the domain of materials science and engineering. These materials intelligence ecosystems are being shaped by several independent developments. Machine learning (ML) algorithms and extant materials data are utilized to create surrogate models of materials properties and performance predictions. Materials data repositories, which fuel such surrogate model development, are mushrooming. Automated data and knowledge capture from the literature (to populate data repositories) using natural language processing approaches is being explored. The design of materials that meet target property requirements and of synthesis steps to create target materials appear to be within reach, either by closed-loop active-learning strategies or by inverting the prediction pipeline using advanced generative algorithms. AI and ML concepts are also transforming the computational and physical laboratory infrastructural landscapes used to create materials data in the first place. Surrogate models that can outstrip physics-based simulations (on which they are trained) by several orders of magnitude in speed while preserving accuracy are being actively developed. Automation, autonomy and guided high-throughput techniques are imparting enormous efficiencies and eliminating redundancies in materials synthesis and characterization. The integration of the various parts of the burgeoning ML landscape may lead to materials-savvy digital assistants and to a human-machine partnership that could enable dramatic efficiencies, accelerated discoveries and increased productivity. Here, we review these emergent materials intelligence ecosystems and discuss the imminent challenges and opportunities. The materials research landscape is being transformed by the infusion of approaches based on machine learning. This Review discusses the emerging materials intelligence ecosystems and the potential of human-machine partnerships for fast and efficient virtual materials screening, development and discovery.
Quantum neural network architectures that have little to no inductive biases are known to face trainability and generalization issues. Inspired by a similar problem, recent breakthroughs in machine learning address this challenge by creating models encoding the symmetries of the learning task. This is materialized through the usage of equivariant neural networks the action of which commutes with that of the symmetry. In this work, we import these ideas to the quantum realm by presenting a comprehensive theoretical framework to design equivariant quantum neural networks (EQNNs) for essentially any relevant symmetry group. We develop multiple methods to construct equivariant layers for EQNNs and analyze their advantages and drawbacks. Our methods can find unitary or general equivariant quantum channels efficiently even when the symmetry group is exponentially large or continuous. As a special implementation, we show how standard quantum convolutional neural networks (QCNNs) can be generalized to group-equivariant QCNNs where both the convolution and pooling layers are equivariant to the symmetry group. We then numerically demonstrate the effectiveness of a S U ( 2 ) -equivariant QCNN over symmetry-agnostic QCNN on a classification task of phases of matter in the bond-alternating Heisenberg model. Our framework can be readily applied to virtually all areas of quantum machine learning. Lastly, we discuss about how symmetry-informed models such as EQNNs provide hopes to alleviate central challenges such as barren plateaus, poor local minima, and sample complexity. Published by the American Physical Society 2024
Exploring planetary environments is central to NASA's missions and goals. A new computing technology called Virtual Reality has much to offer in support of planetary exploration. This technology augments and extends human presence within computer-generated and remote spatial environments. Historically, NASA has been a leader in many of the fundamental concepts and technologies that comprise Virtual Reality. Indeed, Ames Research Center has a central role in the development of this rapidly emerging approach to using computers. This ground breaking work has inspired researchers in academia, industry, and the military. Further, NASA's leadership in this technology has spun off new businesses, has caught the attention of the international business community, and has generated several years of positive international media coverage. In the future, Virtual Reality technology will enable greatly improved human-machine interactions for more productive planetary surface exploration. Perhaps more importantly, Virtual Reality technology will democratize the experience of planetary exploration and thereby broaden understanding of, and support for, this historic enterprise.
This presentation, part of the JISEA 10th Annual (Virtual) Meeting, was part of the panel discussion on machine learning and artificial intelligence for energy solutions.
A high-energy muon collider with center-of-mass energy around and above 10 TeV is also a vector boson fusion (VBF) machine, due to the significant virtual electroweak (EW) gauge boson content of high-energy muon beams. This feature, together with the clean environment, makes it an ideal collider to search for TeV-scale axion-like particles (ALP) coupling to Standard Model EW gauge bosons, which current and other future colliders have limited sensitivities to. We present detailed analyses of heavy ALP searches in both the VBF and associated production channels at a muon collider with different running benchmarks. We also show projected constraints on the ALP couplings in the effective field theory, including an operator with its coefficient not determined by the mixed Peccei-Quinn anomaly. We demonstrate that a muon collider could probe new ALP parameter space and push the sensitivities of the couplings between the ALP and EW gauge bosons by one order of magnitude compared to HL-LHC. The projected limits and search strategies for ALPs could also be applied to other types of resonances coupling to EW gauge bosons.
Multireference (MR) diagnostics are common tools for identifying strongly correlated electronic structure that makes single-reference (SR) methods (e.g., density functional theory or DFT) insufficient for accurate property prediction. However, MR diagnostics typically require computationally demanding correlated wave function theory (WFT) calculations, and diagnostics often disagree or fail to predict MR effects on properties. To overcome these challenges, we introduce a semi-supervised machine learning (ML) approach with virtual adversarial training (VAT) of an MR classifier using 15 WFT and DFT MR diagnostics as inputs. In semi-supervised learning, only the most extreme SR or MR points are labeled, and the remaining point labels are learned. The resulting VAT model outperforms the alternatives, as quantified by the distinct property distributions of SR- and MR-classified molecules. Additionally, to reduce the cost of generating inputs to the VAT model, we leverage the VAT model’s robustness to noisy inputs by replacing WFT MR diagnostics with regression predictions in an MR decision engine workflow that preserves excellent performance. We demonstrate the transferability of our approach to larger molecules and those with distinct chemical composition from the training set. This MR decision engine demonstrates promise as a low-cost, high-accuracy approach to the automatic detection of strong correlation for predictive high-throughput screening.
Radiographic imaging and tomography (RadIT) come in many flavors such as X-ray imaging and tomography (IT), proton IT, neutron IT, muon IT, neutrino IT, etc. We identify five RadIT themes: Physics, Sources, Detectors, Methods, and Data Science, which are integral parts of image interpretation and three-dimensional (3D) tomographic reconstruction. Traditionally, RadIT have been driven by medicine, non-destructive testing, material sciences, and security applications. The latest thrusts of growth come from automation, machine vision, additive manufacturing and virtual reality (‘metaverse’). The five RadIT themes parallel their counterparts in optical IT. Synergies among different forms of RadIT and with optical IT motivate further advances towards multi-modal IT and quantum IT.