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.
Systems and methods for classifying and sorting of plastic materials utilizing a vision system and one or more sensor systems, which may implement a machine learning system in order to identify or classify each of the materials, which may then be sorted into separate groups based on such an identification or classification.
The NASA VEGetation Workbench (VEG) is a knowledge based system that infers vegetation characteristics from reflectance data. VEG is described in detail in several references. The first generation version of VEG was extended. In the first year of this contract, an interface to a file of unknown cover type data was constructed. An interface that allowed the results of VEG to be written to a file was also implemented. A learning system that learned class descriptions from a data base of historical cover type data and then used the learned class descriptions to classify an unknown sample was built. This system had an interface that integrated it into the rest of VEG. The VEG subgoal PROPORTION.GROUND.COVER was completed and a number of additional techniques that inferred the proportion ground cover of a sample were implemented. This work was previously described. The work carried out in the second year of the contract is described. The historical cover type database was removed from VEG and stored as a series of flat files that are external to VEG. An interface to the files was provided. The framework and interface for two new VEG subgoals that estimate the atmospheric effect on reflectance data were built. A new interface that allows the scientist to add techniques to VEG without assistance from the developer was designed and implemented. A prototype Help System that allows the user to get more information about each screen in the VEG interface was also added to VEG.
The NASA VEGetation Workbench (VEG) is a knowledge based system that infers vegetation characteristics from reflectance data. The report describes the extensions that have been made to the first generation version of VEG. An interface to a file of unkown cover type data has been constructed. An interface that allows the results of VEG to be written to a file has been implemented. A learning system that learns class descriptions from a data base of historical cover type data and then uses the learned class descriptions to classify an unknown sample has been built. This system has an interface that integrates it into the rest of VEG. The VEG subgoal PROPORTION.GROUND.COVER has been completed and a number of additional techniques that infer the proportion ground cover of a sample have been implemented.
Frequency control from photovoltaic (PV) power plants has great potential to address the frequency response challenge of the power system with high penetrations of renewable generation. Using model-based approaches to determine the optimal PV headroom reserve, however, requires significant online computation and is intractable for an interconnection level system. This paper proposes a machine learning based strategy, that is suitable for real-time operation, to determine the optimal PV reserve for frequency control. The proposed machine learning algorithm is trained and tested on 1,987 offline simulations of a 60% renewable penetration Western Electricity Coordinating Council (WECC) system. Furthermore, the proposed reserve determination strategy is applied on a realistic 1-day operation profile of the WECC system and demonstrates a savings of more than 40% PV headroom compared to a conservative approach. It is evident that the proposed strategy can efficiently and effectively determine the optimal PV frequency control reserve for realistic interconnection systems.
Standard deep learning algorithms require differentiating large nonlinear networks, a process that is slow and power-hungry. Electronic contrastive local learning networks (CLLNs) offer potentially fast, efficient, and fault-tolerant hardware for analog machine learning, but existing implementations are linear, severely limiting their capabilities. These systems differ significantly from artificial neural networks as well as the brain, so the feasibility and utility of incorporating nonlinear elements have not been explored. Here, we introduce a nonlinear CLLN—an analog electronic network made of self-adjusting nonlinear resistive elements based on transistors. We demonstrate that the system learns tasks unachievable in linear systems, including XOR (exclusive or) and nonlinear regression, without a computer. We find our decentralized system reduces modes of training error in order (mean, slope, curvature), similar to spectral bias in artificial neural networks. The circuitry is robust to damage, retrainable in seconds, and performs learned tasks in microseconds while dissipating only picojoules of energy across each transistor. This suggests enormous potential for fast, low-power computing in edge systems like sensors, robotic controllers, and medical devices, as well as manufacturability at scale for performing and studying emergent learning.
In rule-based AI planning, expert, and learning systems, it is often the case that the left-hand-sides of the rules must be repeatedly compared to the contents of some 'working memory'. The traditional approach to solve such a 'match phase problem' for production systems is to use the Rete Match Algorithm. Here, a new technique using a multilayer perceptron, a particular artificial neural network model, is presented to solve the match phase problem for rule-based AI systems. A syntax for premise formulas (i.e., the left-hand-sides of the rules) is defined, and working memory is specified. From this, it is shown how to construct a multilayer perceptron that finds all of the rules which can be executed for the current situation in working memory. The complexity of the constructed multilayer perceptron is derived in terms of the maximum number of nodes and the required number of layers. A method for reducing the number of layers to at most three is also presented.
Honeywell Systems and Research Center developed and demonstrated an active 35 GHz Radar Imaging system as part of the FAA/USAF/Industry sponsored Synthetic Vision System Technology Demonstration (SVSTD) Program. The objectives of this presentation are to provide a general overview of flight test results, a system level perspective that encompasses the efforts of the SVSTD and Augmented VIsual Display (AVID) programs, and more importantly, provide the AVID workshop participants with Honeywell's perspective on the lessons that were learned from the SVS flight tests. One objective of the SVSTD program was to explore several known system issues concerning radar imaging technology. The program ultimately resolved some of these issues, left others open, and in fact created several new concerns. In some instances, the interested community has drawn improper conclusions from the program by globally attributing implementation specific issues to radar imaging technology in general. The motivation for this presentation is therefore to provide AVID researchers with a better understanding of the issues that truly remain open, and to identify the perceived issues that are either resolved or were specific to Honeywell's implementation.
Power system emergency control is generally regarded as the final safety net for grid security and resiliency. Existing emergency control schemes are usually designed off-line based on either the conceived “worst” case scenarios or a few typical operation scenarios. These schemes are facing significant adaptiveness and robustness issues as increasing uncertainties and variations occur in modern electrical grids. To address these challenges, for the first time, this paper proposes a novel adaptive emergency control scheme using deep reinforcement learning (DRL), by leveraging the high-dimensional feature extraction and non-linear generalization capabilities DRL has for complex systems with high-dimensional variations. Furthermore, an open-source platform named DeepGrid has been designed for the first time to assist the DRL development and benchmarking processes in power system emergency control. Details of the platform, DRL, and emergency control schemes that use dynamic braking or under-voltage load shedding are presented. Extensive case studies performed in both two-area four-machine system and IEEE 39-Bus system have demonstrated the excellent performance and robustness of the proposed schemes.
Federated learning (FL) is increasingly deployed on accelerator-rich high-performance computing (HPC) systems, yet the system-level energy cost of privacy-aware FL remains poorly understood, particularly across heterogeneous networking and server-placement options. We present a measurement-driven study of energy–performance trade-offs for FL on GH200-class nodes across three deployment configurations: CPU-Ethernet, CPU-InfiniBand (RDMA-capable), and a DPU-hosted FL server over InfiniBand using a BlueField-3 SmartNIC/DPU. Using NVIDIA FLARE (NVFLARE), we align node-level power telemetry with per-round timing extracted from NVFLARE logs to quantify time-to-solution (TTS), energy-to-solution (ETS), energy-delay product (EDP), and synchronization behavior for three transformer models (ALBERT, DistilBERT, BERT), trained with and without differential privacy (DP). We find that interconnect choice is the dominant driver of runtime and energy: host-managed InfiniBand consistently reduces communication overhead versus Ethernet, yielding lower TTS/ETS/EDP. In contrast, in our NVFLARE deployment, placing the FL server on the DPU does not consistently match CPU-InfiniBand performance and can be slower—especially for larger models—highlighting that server placement alone is not sufficient to guarantee end-to-end gains. Finally, under our fixed-round protocol, DP increases per-round cost and runtime variance; ETS increases largely in proportion to TTS because average node power remains relatively stable across configurations.
Deep reinforcement learning (DRL) algorithms have successfully solved many challenging problems in various power system control scenarios. However, their decision-making process is usually regarded as black-boxes. Furthermore, how DRL models interact with human intelligence remains an open problem. Thus, this paper proposes a policy extraction framework to extract a complex DRL model into an explainable policy. This framework includes three parts: 1) DRL training and data generation. We train an agent for a specific control task and generate data, which contains the control policy of the agent. 2) Policy extraction. We propose an information gain rate based weighted oblique decision tree (IGR-WODT) for DRL policy extraction. 3) Policy evaluation. We define three metrics to evaluate the performance of the proposed approach. A case study for the under-voltage load shedding problem shows that the IGR-WODT presents a performance enhancement compared with DRL, weighted oblique decision tree, and univariate decision tree. The proposed policy extraction method could provide an intuitive explanation of the neural network decision-making process to the dispatchers when making final decisions on power grid operation. Also, the resulted rule-based controller could replace the deep neural network-based controller in many field edge devices with limited computing resources, providing comparable performance.
A considerable amount of work on the development of fuzzy logic algorithms and application to space related control problems has been done at the Johnson Space Center (JSC) over the past few years. Particularly, guidance control systems for space vehicles during proximity operations, learning systems utilizing neural networks, control of data processing during rendezvous navigation, collision avoidance algorithms, camera tracking controllers, and tether controllers have been developed utilizing fuzzy logic technology. Several other areas in which fuzzy sets and related concepts are being considered at JSC are diagnostic systems, control of robot arms, pattern recognition, and image processing. It has become evident, based on the commercial applications of fuzzy technology in Japan and China during the last few years, that this technology should be exploited by the government as well as private industry for energy savings.
Reliability and stability of the electric grid today has depended upon operations of the grid which include the protective relay. Today, the electricity sector faces new challenges with the shift of generation resource characteristics away from the traditional “big iron” generation to inverter-based resources (IBR) which shift the physics and assumption used in grid operation and protection. These changing conditions represent new challenges for protective relays (identification of faults) and increased challenges for protection engineers (correct settings and configuration, reduction of mis-operations), both issues recognized in research and industry. Finding new approaches to reduce mis-operations in relaying and new approaches to fault identification is critical to grid operations. Using today’s modern technology of embedded systems, edge computing, machine learning (ML), and communications we can help address challenges and augment and improve on existing power system operations methodologies.
The differences between the terms learning curve and improvement curve are noted, as well as the differences between the Wright system and the Crawford system. Learning curve computational techniques were reviewed along with a method to arrive at a composite learning curve for a system given detail curves either by the functional techniques classification or simply categorized by subsystem. Techniques are discussed for determination of the theoretical first unit (TFU) cost using several of the currently accepted methods. Sometimes TFU cost is referred to as simply number one cost. A tabular presentation of the various learning curve slope values is given. A discussion of the various trends in the application of learning/improvement curves and an outlook for the future are presented.
We develop a methodology based on Bayesian inference over Probabilistic Graphical Models (PGMs) to understand and quantify risk in solar-centered grids using targeted measurements and learned system behavior. Being non-prescriptive but, rather, able to infer system behavior and, ultimately, address risk queries from data, our machine learning-type paradigm is tailored for diverse topologies and threat scenarios often associated with distributed energy generation and photovoltaic distributed energy resources (PV-DERs) in particular. We describe algorithmic processes for: (i) learning the structure of PGMs that result from attack-prone PV-DER-proliferated distribution systems, (ii) quantifying cause-effect relationships, and (iii) evaluating risk queries based on diverse evidence. The contributions are illustrated on a residential grid subject to output impairment attacks on its PV-DER infrastructure.
This report contains a complete description of a learning control system designed for the F8-DFBW aircraft. The system is parameter-adaptive with the additional feature that it 'learns' the variation of the control system gains needed over the flight envelope. It, thus, generates and modifies its gain schedule when suitable data are available. The report emphasizes the novel learning features of the system: the forms of representation of the flight envelope and the process by which identified parameters are used to modify the gain schedule. It contains data taken during piloted real-time 6 degree-of-freedom simulations that were used to develop and evaluate the system.
Deep learning (DL) workloads and their performance at scale are becoming important factors to consider as we design, develop and deploy next-generation high-performance computing systems. Since DL applications rely heavily on DL frameworks and underlying compute (CPU/GPU) stacks, it is essential to gain a holistic understanding from compute kernels, models, and frameworks of popular DL stacks, and to assess their impact on science-driven, mission-critical applications. At Oak Ridge Leadership Computing Facility (OLCF), we employ a set of micro and macro DL benchmarks established through the Collaboration of Oak Ridge, Argonne, and Livermore (CORAL) to evaluate the AI readiness of our next-generation supercomputers. In this paper, we present our early observations and performance benchmark comparisons between the Nvidia V100 based Summit system with its CUDA stack and an AMD MI100 based testbed system with its ROCm stack. We take a layered perspective on DL benchmarking and point to opportunities for future optimizations in the technologies that we consider.
An Intelligent Workbench (VEG) was developed for the systematic study of remotely sensed optical data from vegetation. A goal of the remote sensing community is to infer the physical and biological properties of vegetation cover (e.g. cover type, hemispherical reflectance, ground cover, leaf area index, biomass, and photosynthetic capacity) using directional spectral data. VEG collects together, in a common format, techniques previously available from many different sources in a variety of formats. The decision as to when a particular technique should be applied is nonalgorithmic and requires expert knowledge. VEG has codified this expert knowledge into a rule-based decision component for determining which technique to use. VEG provides a comprehensive interface that makes applying the techniques simple and aids a researcher in developing and testing new techniques. VEG also provides a classification algorithm that can learn new classes of surface features. The learning system uses the database of historical cover types to learn class descriptions of one or more classes of cover types.
Attitude determination has been considered as a permanent topic of active research and perhaps remaining as a forever-lasting interest for spacecraft system designers. Its role is to provide a reference for controls such as pointing the directional antennas or solar panels, stabilizing the spacecraft or maneuvering the spacecraft to a new orbit. Least Square Estimation (LSE) technique was utilized to provide attitude determination for the Nimbus 6 and G. Despite its poor performance (estimation accuracy consideration), LSE was considered as an effective and practical approach to meet the urgent need and requirement back in the 70's. One reason for this poor performance associated with the LSE scheme is the lack of dynamic filtering or 'compensation'. In other words, the scheme is based totally on the measurements and no attempts were made to model the dynamic equations of motion of the spacecraft. We propose an adaptive filtering approach which employs a bank of Kalman filters to perform robust attitude estimation. The proposed approach, whose architecture is depicted, is essentially based on the latest proof on the interactive multiple model design framework to handle the unknown of the system noise characteristics or statistics. The concept fundamentally employs a bank of Kalman filter or submodel, instead of using fixed values for the system noise statistics for each submodel (per operating condition) as the traditional multiple model approach does, we use an on-line dynamic system noise identifier to 'identify' the system noise level (statistics) and update the filter noise statistics using 'live' information from the sensor model. The advanced noise identifier, whose architecture is also shown, is implemented using an advanced system identifier. To insure the robust performance for the proposed advanced system identifier, it is also further reinforced by a learning system which is implemented (in the outer loop) using neural networks to identify other unknown quantities such as spacecraft dynamics parameters, gyro biases, dynamic disturbances, or environment variations.