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.
This presentation was presented to SWS Data Mining and Prediction RTT subgroup on behalf of DMTA ML/AI processes RTT subgroup. The slides presented a review of ML airport surface model and showed validation results at KDFW.
Resins filled with granules formed to final size and shape in new process. New vented compression molding method takes less time than conventional method. Vacuum-bagging and machining stages eliminated. Part emerges from mold in final size and shape. Suitable for making variety of parts with simple shapes, vented compression molding takes less time and cost less than older processes.
Arc jet Computer Vision (ArcjetCV) is a software application built to automate analysis of arc jet ground test video footage. This includes tracking material recession and the shock-material standoff distance. This provides a new capability to resolve and validate new physics associated with non-linear processes. This is an essential step to reduce testing, modeling, and validation uncertainties for heatshield material performance. ArcjetCV uses several types of machine learning (convolutional neural net: CNN, decision tree: DT, k-means unsupervised clustering: KM) to automate the video processing pipeline. These include inferring the start/stop of time segments of interest (1D CNN), measuring the time-dependent 2D recession of the material samples (2D CNN, DT), measuring the time-dependent shock standoff distance (2D CNN, DT), and post-processing cleaning of the recession data (KM). The software also provides a graphical user interface for ease of use. The results of using this tool on arc jet videos show non-linear time-dependent effects can be important for certain materials and characterizing certain failure modes.
Arc jet Computer Vision (ArcjetCV) is a software application built to automate analysis of arc jet ground test video footage. This includes tracking material recession and the shock-material standoff distance. This provides a new capability to resolve and validate new physics associated with non-linear processes. This is an essential step to reduce testing, modeling, and validation uncertainties for heatshield material performance. ArcjetCV uses several types of machine learning (convolutional neural net: CNN, decision tree: DT, k-means unsupervised clustering: KM) to automate the video processing pipeline. These include inferring the start/stop of time segments of interest (1D CNN), measuring the time-dependent 2D recession of the material samples (2D CNN, DT), measuring the time-dependent shock standoff distance (2D CNN, DT), and post-processing cleaning of the recession data (KM). The software also provides a graphical user interface for ease of use. The results of using this tool on arc jet videos show non-linear time-dependent effects can be important for certain materials and characterizing certain failure modes.
Multi-messenger astrophysics is a fast-growing, interdisciplinary field that combines data, which vary in volume and speed of data processing, from many different instruments that probe the Universe using different cosmic messengers: electromagnetic waves, cosmic rays, gravitational waves and neutrinos. In this Expert Recommendation, we review the key challenges of real-time observations of gravitational wave sources and their electromagnetic and astroparticle counterparts, and make a number of recommendations to maximize their potential for scientific discovery. These recommendations refer to the design of scalable and computationally efficient machine learning algorithms; the cyber-infrastructure to numerically simulate astrophysical sources, and to process and interpret multi-messenger astrophysics data; the management of gravitational wave detections to trigger real-time alerts for electromagnetic and astroparticle follow-ups; a vision to harness future developments of machine learning and cyber-infrastructure resources to cope with the big-data requirements; and the need to build a community of experts to realize the goals of multi-messenger astrophysics.
The System Diagnostic Builder (SDB) is an automated software verification and validation tool using state-of-the-art Artificial Intelligence (AI) technologies. The SDB is used extensively by project BURKE at NASA-JSC as one component of a software re-engineering toolkit. The SDB is applicable to any government or commercial organization which performs verification and validation tasks. The SDB has an X-window interface, which allows the user to 'train' a set of rules for use in a rule-based evaluator. The interface has a window that allows the user to plot up to five data parameters (attributes) at a time. Using these plots and a mouse, the user can identify and classify a particular behavior of the subject software. Once the user has identified the general behavior patterns of the software, he can train a set of rules to represent his knowledge of that behavior. The training process builds rules and fuzzy sets to use in the evaluator. The fuzzy sets classify those data points not clearly identified as a particular classification. Once an initial set of rules is trained, each additional data set given to the SDB will be used by a machine learning mechanism to refine the rules and fuzzy sets. This is a passive process and, therefore, it does not require any additional operator time. The evaluation component of the SDB can be used to validate a single software system using some number of different data sets, such as a simulator. Moreover, it can be used to validate software systems which have been re-engineered from one language and design methodology to a totally new implementation.
Arc jet Computer Vision (ArcjetCV) is a software application built to automate analysis of arc jet ground test video footage. This includes tracking material recession, sting arm motion, and the shock-material standoff distance. This provides a new capability to resolve and validate new physics associated with non-linear processes. This is an essential step to reduce testing, modeling, and validation uncertainties for heatshield material performance. ArcjetCV uses several types of machine learning (convolutional neural net: CNN, decision tree: DT, k-means unsupervised clustering: KM) to automate the video processing pipeline. These include inferring the start/stop of time segments of interest (1D CNN), measuring the time-dependent 2D recession of the material samples (2D CNN, DT), measuring the time-dependent shock standoff distance (2D CNN, DT), and post-processing cleaning of the recession data (KM). The software also provides a graphical user interface for ease of use. The results of using this tool on arc jet videos show non-linear time-dependent effects can be important for certain materials.
Arc jet Computer Vision (ArcjetCV) is a software application built to automate analysis of arc jet ground test video footage. This includes tracking material recession, sting arm motion, and the shock-material standoff distance. This provides a new capability to resolve and validate new physics associated with non-linear processes. This is an essential step to reduce testing, modeling, and validation uncertainties for heatshield material performance. ArcjetCV uses several types of machine learning (convolutional neural net: CNN, decision tree: DT, k-means unsupervised clustering: KM) to automate the video processing pipeline. These include inferring the start/stop of time segments of interest (1D CNN), measuring the time-dependent 2D recession of the material samples (2D CNN, DT), measuring the time-dependent shock standoff distance (2D CNN, DT), and post-processing cleaning of the recession data (KM). The software also provides a graphical user interface for ease of use. The results of using this tool on arc jet videos show non-linear time-dependent effects can be important for certain materials.
The successful development of an advanced powder metallurgy disk alloy, ME3, was initiated in the NASA High Speed Research/Enabling Propulsion Materials (HSR/EPM) Compressor/Turbine Disk program in cooperation with General Electric Engine Company and Pratt & Whitney Aircraft Engines. This alloy was designed using statistical screening and optimization of composition and processing variables to have extended durability at 1200 F in large disks. Disks of this alloy were produced at the conclusion of the program using a realistic scaled-up disk shape and processing to enable demonstration of these properties. The objective of the Ultra-Efficient Engine Technologies disk program was to assess the mechanical properties of these ME3 disks as functions of temperature in order to estimate the maximum temperature capabilities of this advanced alloy. These disks were sectioned, machined into specimens, and extensively tested. Additional sub-scale disks and blanks were processed and selectively tested to explore the effects of several processing variations on mechanical properties. Results indicate the baseline ME3 alloy and process can produce 1300 to 1350 F temperature capabilities, dependent on detailed disk and engine design property requirements.
Additive manufacturing (AM) has unique process attributes that facilitate the creation of optimized, complex, and unique parts for aerospace applications. However, the multi-scale and complicated building process for AM parts can cause unexpected build conditions that result in microstructural variability within the deposited material. The need for richer datasets and computational modeling capabilities to improve AM process reliability is a consequence of this microstructural variability. A computational approach, referred to as the Additive Manufacturing Moment Measure (AM3), addresses this need by leveraging the way AM processes fuse material in precise, incremental steps. During the layering sequence, each previous step contributes to the condition of the current step at the center of the heat source. The precise steps can be selected to systematically create a time-space point field with co-located machine input and in-situ sensor data. The sequential nature of the AM process, coupled with nearest neighbor calculations, allows for a fully parallel computation for part-scale build profile analysis. The AM3 concept introduced here enables part-scale assessment directly from build files and in-situ process monitoring sensors alike. The AM moment measures were calculated for build point fields and compared with co-located in-situ and ex-situ nondestructive evaluation and optical microscopy observations. These comparisons permit a better understanding of how the sequential process actions can affect the quality of a laser powder bed fusion (LPBF) build. Details of the AM3technique will be discussed and compared to measured LPBF part characteristics. The AM3 results indicate a strong potential to advance the qualification process for aerospace applications.
The report discusses the: Decentralized Control of Markov Decision Processes. Study the complexity of decentralized control of Markov decision processes, and develop algorithms for finding optimal control policies. Scheduling Contract Algorithms. Develop an optimal method for scheduling runs of a contract anytime algorithm (one that takes the deadline as input) in situations where the deadline is unknown, multiple problem instances must be solved, and a multi-processor machine is available. Planetary Rover Control as a Markov Decision Process.Use the Markov decision process framework to formalize and solve problems in planetary rover control. Adaptive Peer Selection. Use reinforcement learning to maximize the expected down-load speed for a client in a peer-to-peer file sharing system.
Additive Manufacturing (AM) is a technology that builds three dimensional objects by adding material layer-upon-layer throughout the fabrication process. The Electrical, Electronic and Electromechanical (EEE) parts packaging group at Marshall Space Flight Center (MSFC) is investigating how various AM and 3D printing processes can be adapted to the microgravity environment of space to enable on demand manufacturing of electronics. The current state-of-the art processes for accomplishing the task of printing electronics through non-contact, direct-write means rely heavily on the process of atomization of liquid inks into fine aerosols to be delivered ultimately to a machine's print head and through its nozzle. As a result of cumulative International Space Station (ISS) research into the behaviors of fluids in zero-gravity, our experience leads us to conclude that the direct adaptation of conventional atomization processes will likely fall short and alternative approaches will need to be explored. In this report, we investigate the development of an alternative approach to atomizing electronic materials by way of thermal atomization, to be used in place of conventional aerosol generation and delivery processes for printing electronics in space.
High-performance computing (HPC) architectures are trending toward manycore paradigms such as graphics processing units (GPUs). Approximately half of the top 100 publicly disclosed supercomputers in the world utilize GPU accelerators for performance. This is in contrast to a decade ago, where there were only a few such machines in the top 100. It is not currently possible to compile and run legacy central processing unit (CPU) software efficiently on GPUs without significant refactoring. Though a number of frameworks offering performance portability exist, none offer a standardized specification that is supported by all major hardware vendors. Additionally, experiences show that obtaining a high percentage of peak performance often requires architecture-specific code. This work details a pragmatic multi-architecture computational fluid dynamics library focused on aerospace problems across the speed range from low subsonic to hypersonic flows involving thermochemical nonequilibrium. A thin abstraction layer above NVIDIA CUDA C++ is utilized, which enables primarily single-source software currently capable of running efficiently on multicore CPUs, NVIDIA GPUs, AMD GPUs, and Intel GPUs. Results on various problems of interest across the speed range are presented and performance is compared between various architectures.
High-performance computing (HPC) architectures are trending toward manycore paradigms such as graphics processing units (GPUs). Approximately half of the top 100 publicly disclosed supercomputers in the world utilize GPU accelerators for performance. This is in contrast to a decade ago, where there were only a few such machines in the top 100. It is not currently possible to compile and run legacy central processing unit (CPU) software efficiently on GPUs without significant refactoring. Though a number of frameworks offering performance portability exist, none offer a standardized specification that is supported by all major hardware vendors. Additionally, experiences show that obtaining a high percentage of peak performance often requires architecture-specific code. This work details a pragmatic multi-architecture computational fluid dynamics library focused on aerospace problems across the speed range from low subsonic to hypersonic flows involving thermochemical nonequilibrium. A thin abstraction layer above NVIDIA CUDA C++ is utilized, which enables primarily single-source software currently capable of running efficiently on multicore CPUs, NVIDIA GPUs, AMD GPUs, and Intel GPUs. Results on various problems of interest across the speed range are presented and performance is compared between various architectures.
An account if given of the variational method of the solution of physically and geometrically nonlinear problems of the theory of heterogeneous slightly curved shells. Examined are the bending and supercritical behavior of plates and conical and spherical cupolas of variable thickness in a temperature field, taking into account the dependence of the elastic parameters on temperature. The bending, stability in general and load-bearing capacity of flexible isotropic elastic-plastic shells with different criteria of plasticity, taking into account compressibility and hardening. The effect of the plastic heterogeneity caused by heat treatment, surface work hardening and irradiation by fast neutron flux is investigated. Some problems of the dynamic behavior of flexible shells are solved. Calculations are performed in high approximations. Considerable attention is given to the construction of a machine algorithm and to the checking of the convergence of iterative processes.
The feasibility of mounting a small operation on the Moon to productively use lunar materials in support of programs such as the solar power satellite is addressed. A cost effective scenario of a small chemical process plant on the surface of the Moon and a small machine shop located in orbit is presented. The mass of the space installation is compared to the projected outputs in 90 days. It is indicated that the system would have the capability of replicating about 90% of its own components and would provide the metals, glasses, and silicon needed for the contruction of 90% to 96% of the mass of one solar power satellite per year.
The machine-readable catalog provides mean data on the old Slettebak system for 6472 stars. The catalog results from the review, analysis and transformation of 11460 data from 102 sources. Star identification, (major catalog number, name if the star has one, or cluster identification, etc.), a man projected rotational velocity, and a list of source references re included. The references are given in a second file included with the catalog when it is distributed on magnetic tape. The contents and/formats of the the data and reference files of the machine-readable catalog are described to enable users to read and process the data.
A methodology was developed a assess the upset susceptibility/reliability of a computer system onboard an aircraft flying through a lightning environment. Upset error modes in a general purpose microprocessor were studied. The upset tests involved the random input of analog transients which model lightning induced signals onto interface lines of an 8080 based microcomputer from which upset error data was recorded. The program code on the microprocessor during tests is designed to exercise all of the machine cycles and memory addressing techniques implemented in the 8080 central processing unit. A statistical analysis is presented in which possible correlations are established between the probability of upset occurrence and transient signal inputs during specific processing states and operations. A stochastic upset susceptibility model for the 8080 microprocessor is presented. The susceptibility of this microprocessor to upset, once analog transients have entered the system, is determined analytically by calculating the state probabilities of the stochastic model.