Search NASA⌕ Search

SEARCH · Search NASA

Results for “machine learning operations”

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 1,657 records · Page 92

Fault Injection for TensorFlow Applications

As machine learning (ML) has seen increasing adoption in safety-critical domains (e.g., autonomous vehicles), the reliability of ML systems has also grown in importance. While prior studies have proposed techniques to enable efficient error-resilience (e.g., selective instruction duplication), a fundamental requirement for realizing these techniques is a detailed understanding of the application’s resilience. In this work, we present TensorFI 1 and TensorFI 2, high-level fault injection (FI) frameworks for TensorFlow-based applications. TensorFI 1 and 2 are able to inject both hardware and software faults in any general TensorFlow 1 and 2 program respectively. Both are configurable FI tools that are flexible, easy to use, and portable. They can be integrated into existing TensorFlow programs to assess their resilience for different fault types (e.g., bit-flips in particular operations or layers). We use the TensorFI 1 and TensorFI 2 to evaluate the resilience of 12 and 10 ML programs written in TensorFlow, including DNNs used in the autonomous vehicle domain. The results give us insights into why some of the models are more resilient. We also measure the performance overheads of the two injectors, and present 4 case studies, two for each tool, to demonstrate their utility.

97 MATHEMATICS AND COMPUTING↗

A Multi-scale, Multi-Model, Machine-Learning Solar Forecasting Technology

The goal of the project was the development and demonstration of a significantly improved solar forecasting technology (short: Watt-sun), which leverages new big data processing technologies and machine-learnt blending between different models and forecast systems. The technology aimed demonstrating major advances in accuracy as measured by existing and new metrics which themselves were developed as part of this project. Finally, the team worked with Independent System Operators (ISOs) and utilities to integrate the forecasts into their operations.

97 MATHEMATICS AND COMPUTING↗

Neuromorphic Computing is Turing-Complete

Neuromorphic computing is a non-von Neumann computing paradigm that performs computation by emulating the human brain. Neuromorphic systems are extremely energy-efficient and known to consume thousands of times less power than CPUs and GPUs. They have the potential to drive critical use cases such as autonomous vehicles, edge computing and internet of things in the future. For this reason, they are sought to be an indispensable part of the future computing landscape. Neuromorphic systems are mainly used for spike-based machine learning applications, although there are some non-machine learning applications in graph theory, differential equations, and spike-based simulations. These applications suggest that neuromorphic computing might be capable of general-purpose computing. However, general-purpose computability of neuromorphic computing has not been established yet. In this work, we prove that neuromorphic computing is Turing-complete and therefore capable of general-purpose computing. Specifically, we present a model of neuromorphic computing, with just two neuron parameters (threshold and leak), and two synaptic parameters (weight and delay). We devise neuromorphic circuits for computing all the μ-recursive functions (i.e., constant, successor and projection functions) and all the μ-recursive operators (i.e., composition, primitive recursion and minimization operators). Given that the μ-recursive functions and operators are precisely the ones that can be computed using a Turing machine, this work establishes the Turing-completeness of neuromorphic computing.

Date, Prasanna↗

A dimensionality reduction algorithm for mapping tokamak operational regimes using a variational autoencoder (VAE) neural network

Abstract A variational autoencoder (VAE) is a type of unsupervised neural network which is able to learn meaningful data representations in a reduced dimensional space. We present an application of VAE in identifying the operational stability boundary of tokamak plasma discharges. This model was implemented using a dataset of over 3000 discharges from the high beta tokamak-extended pulse (HBT-EP) device. We found the VAE model to be capable of forming a continuous low-dimensional operational space map and identifying the operational boundaries using a specified warning time window. By projecting the operational parameters onto the same reduced space, this provides an intuitive way for the machine operator or an automated control system to perform disruption avoidance using a relevant control actuator as a discharge approaches a boundary. Pre-programmed GPU control experiments were conducted to demonstrate this control technique using HBT-EP’s saddle control coils as a horizontal position actuator, showing the ability to avoid the oncoming disruptive event and extend the duration of the discharge.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Hot, cold, or just right? An infrared biometric sensor to improve occupant comfort and reduce overcooling in buildings via closed-loop control

To improve occupant comfort and save energy in buildings, we have developed a closed-loop air conditioning (AC) sensor-controller that predicts occupant thermal sensation from the thermographic measurement of skin temperature distribution, then uses this information to reduce overcooling (cooling-energy overuse that discomforts occupants) by regulating AC output. Taking measures to protect privacy, it combines thermal-infrared (TIR) and color (visible spectrum) cameras with machine vision to measure the skin-surface temperature profile. Since the human thermoregulation system uses skin blood flow to maintain thermoneutrality, the distribution of skin temperature can be used to predict warm, neutral, and cool thermal states. We conducted a series of human-subject thermal-sensation trials in cold-to-hot environments, measuring skin temperatures and recording thermal sensation votes. We then trained random-forest classification machine-learning models (classifiers) to estimate thermal sensation from skin temperatures or skin-temperature differences. The estimated thermal sensation was input to a proportional integral (PI) control algorithm for the AC, targeting a sensation level between neutral and warm. Our sensor-controller includes a sensor assembly, server software, and client software. The server software orients the cameras and transmits images to the client software, which in turn assesses occupant skin temperature distribution, estimates occupant thermal sensation, and controls AC operation. A demonstration conducted in a conference room in an office building near Houston, TX showed that our system reduced overcooling, decreasing AC load by 42% when the room was occupied while improving occupant comfort (fraction of “comfortable” votes) by 15 percentage points.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

De novo design of programmable inducible promoters

Abstract Ligand-responsive allosteric transcription factors (aTF) play a vital role in genetic circuits and high-throughput screening because they transduce biochemical signals into gene expression changes. Programmable control of gene expression from aTF-regulated promoter is important because different downstream effector genes function optimally at different expression levels. However, tuning gene expression of native promoters is difficult due to complex layers of homeostatic regulation encoded within them. We engineered synthetic promoters de novo by embedding operator sites with varying affinities and radically reshaped binding preferences within a minimal, constitutive Escherichia coli promoter. Multiplexed cell-based screening of promoters for three TetR-like aTFs generated with this approach gave rich diversity of gene expression levels, dynamic ranges and ligand sensitivities and were 50- to 100-fold more active over their respective native promoters. Machine learning on our dataset revealed that relative position of the core motif and bases flanking the core motif play an important role in modulating induction response. Our generalized approach yields customizable and programmable aTF-regulated promoters for engineering cellular pathways and enables the discovery of new small molecule biosensors.

59 BASIC BIOLOGICAL SCIENCES↗

SU-F-R-22: Malignancy Classification for Small Pulmonary Nodules with Radiomics and Logistic Regression

Purpose: We conducted a retrospective study of Radiomics research for classifying malignancy of small pulmonary nodules. A machine learning algorithm of logistic regression and open research platform of Radiomics, IBEX (Imaging Biomarker Explorer), were used to evaluate the classification accuracy. Methods: The training set included 100 CT image series from cancer patients with small pulmonary nodules where the average diameter is 1.10 cm. These patients registered at Chang Gung Memorial Hospital and received a CT-guided operation of lung cancer lobectomy. The specimens were classified by experienced pathologists with a B (benign) or M (malignant). CT images with slice thickness of 0.625 mm were acquired from a GE BrightSpeed 16 scanner. The study was formally approved by our institutional internal review board. Nodules were delineated and 374 feature parameters were extracted from IBEX. We first used the t-test and p-value criteria to study which feature can differentiate between group B and M. Then we implemented a logistic regression algorithm to perform nodule malignancy classification. 10-fold cross-validation and the receiver operating characteristic curve (ROC) were used to evaluate the classification accuracy. Finally hierarchical clustering analysis, Spearman rank correlation coefficient, and clustering heat map were used to further study correlation characteristics among different features. Results: 238 features were found differentiable between group B and M based on whether their statistical p-values were less than 0.05. A forward search algorithm was used to select an optimal combination of features for the best classification and 9 features were identified. Our study found the best accuracy of classifying malignancy was 0.79±0.01 with the 10-fold cross-validation. The area under the ROC curve was 0.81±0.02. Conclusion: Benign nodules may be treated as a malignant tumor in low-dose CT and patients may undergo unnecessary surgeries or treatments. Our study may help radiologists to differentiate nodule malignancy for low-dose CT.

60 APPLIED LIFE SCIENCES↗

UAV Research at NASA Langley: Towards Safe, Reliable, and Autonomous Operations

Unmanned Aerial Vehicles (UAV) are fundamental components in several aspects of research at NASA Langley, such as flight dynamics, mission-driven airframe design, airspace integration demonstrations, atmospheric science projects, and more. In particular, NASA Langley Research Center (Langley) is using UAVs to develop and demonstrate innovative capabilities that meet the autonomy and robotics challenges that are anticipated in science, space exploration, and aeronautics. These capabilities will enable new NASA missions such as asteroid rendezvous and retrieval (ARRM), Mars exploration, in-situ resource utilization (ISRU), pollution measurements in historically inaccessible areas, and the integration of UAVs into our everyday lives all missions of increasing complexity, distance, pace, and/or accessibility. Building on decades of NASA experience and success in the design, fabrication, and integration of robust and reliable automated systems for space and aeronautics, Langley Autonomy Incubator seeks to bridge the gap between automation and autonomy by enabling safe autonomous operations via onboard sensing and perception systems in both data-rich and data-deprived environments. The Autonomy Incubator is focused on the challenge of mobility and manipulation in dynamic and unstructured environments by integrating technologies such as computer vision, visual odometry, real-time mapping, path planning, object detection and avoidance, object classification, adaptive control, sensor fusion, machine learning, and natural human-machine teaming. These technologies are implemented in an architectural framework developed in-house for easy integration and interoperability of cutting-edge hardware and software.

Davila, Carlos G.↗

Prediction of Pushback Times and Ramp Taxi Times for Departures at Charlotte Airport

When optimizing the takeoff sequence and schedule for departures at busy airports, it is important to accurately predict the taxi times from gate to runway because those are used to calculate the earliest possible takeoff times. Several airports like Charlotte Douglas International Airport show relatively long taxi times inside the ramp area with large variations, with respect to the travel times in the airport movement area. Also, the pushback process times have not been accurately modeled so far mainly due to the lack of accurate data. The recent deployment of the integrated arrival, departure, and surface traffic management system at Charlotte airport by NASA enables more accurate flight data in the airport surface operations to be obtained. Taking advantage of this system, actual pushback times and ramp taxi times from historical flight data at this airport are analyzed. Based on the analysis, a simple, data-driven prediction model is introduced for estimating pushback times and ramp transit times of individual departure flights. To evaluate the performance of this prediction model, several machine learning techniques are also applied to the same dataset. The prediction results show that the data-driven prediction model is as good as the machine learning algorithms when comparing various prediction performance metrics.

Lee, Hanbong↗

Machine Learning Methods for Estimating Propeller Source Noise Spheres

In this work, several neural network function approximations are compared for inter- polating, storing, and sampling acoustic source spheres with applications to propeller noise estimation. These methods are compared using an acoustic model of the three bladed GL-10 propeller at different flight conditions, with training data generated using NASA’s ANOPP-PAS module. The source spheres used to train the networks capture the tonal propeller noise due to both the blade thickness and loading. This tonal noise prediction method allows the vehicle noise to be estimated for auralization and acoustic control. Three radial basis function neural network architectures are compared in this work. The first two networks directly estimate the parameters of the source sphere at different flight conditions but differ in the number of layers used. The third network estimates the parameters of the source sphere using a weighted combination of spherical basis functions. These networks are trained on numerically generated source spheres, with operating points given in terms of the propeller rotation rate, freestream speed, and propeller angle of attack. The performance of the neural network is determined using a validation dataset of withheld data points. This performance is quantified in terms of the approximation error, training time, and sample time. The third network, which estimates the weights of the spherical basis functions, performs the best in both average and maximum approximation errors in all cases. This network’s worst case performance is 5.6 % relative dif- ference of a model parameter associated with acoustic pressure. The direct estimation network with a single layer has the worst approximation error in all cases. Additionally, the spherically defined network has the slowest sample time at 0.05 seconds per thousand points. Both direct estimation methods produce a thousand sample points in approximately 0.001 seconds.

Acoustics↗

Jefferson Laboratory C100 Superconducting Radio-Frequency Cavity Fault Data, 2020

The dataset was created to train machine learning models for the task of identifying the (1) cavity and (2) fault type from C100-type cryomodules at the Thomas Jefferson National Accelerator Facility (Jefferson Lab), thereby replacing the time-consuming efforts of a subject matter expert. Superconducting radio-frequency (SRF) cavity trips represent a significant source of accelerator downtime. Real-time – rather than post-mortem – identification of the offending cavity and classification of the fault type would give control room operators valuable feedback for corrective action planning. The anticipated benefit is increased beam-on-target time for users and provides performance metrics that can be used to improve future cavity designs. A series of 17 RF signals are recorded for each of the 8 cavities in a C100 cryomodule every time a cavity trips. These time-series signals are written to file using a specially designed data acquisition system. The dataset represents fault events recorded during Continuous Electron Beam Accelerator Facility (CEBAF) beam operations between January 18, 2019 and March 9, 2020. The following filtering steps were applied to collected data; (1) only 4 of the 17 signals per cavity are retained (GMES, GASK, CRFP, DETA2) (2) only events with data from each of the eight cavities in the cryomodule are kept, (3) only events that were sampled at 5 kHz were kept, (4) events from cryomodule 0L04 were neglected, (5) events occurring between February 4, 3PM and February 5, 12PM were neglected. As a result of preprocessing, the dataset is comprised of 2,375 unique events. The full dataset is comprised of three files: features.csv, cavity_labels.csv, fault_labels.csv. Each instance in faults.csv includes a timestamp (“date_time”), a label for the cryomodule which experienced the trip (“zone_label”), and 192 features (“feature_1”, “feature_2”... “feature_192”). The features correspond to 6 autoregressive features for each of 4 signals per cavity for each of the 8 cavities (6 × 4 signals/cavity × 8 cavities/cryomodule = 192). To deal with the large variation of signal amplitudes, time-series standardization via the z-score (standard score) function was applied prior to computing the features. For each instance, there is an associated label for the (1) cavity which faulted first (cavity_labels.csv) and (2) the type of fault that caused the trip (fault_labels.csv). The cavity identification can take values of [0, 1, 2, 3, 4, 5, 6, 7, 8] and the fault type can take values of [‘Microphonics’, ‘Quench_100ms’, ‘Controls_Fault’, ‘E_Quench’, ‘Quench_3ms’, ‘Single_Cav_Turn_Off’ , ‘Heat_Riser_Choke’, ‘Multi_Cav_Turn_Off’].

43 PARTICLE ACCELERATORS↗

A Multi-Sensor Approach for Measuring Bird and Bat Collisions with Offshore Wind Turbines (Final Technical Report)

Collision of birds and bats with wind turbines is a conservation concern for both land-based and offshore wind projects. The fatality rates of birds and bats at land-based turbines are well documented. The measurement strategies on land focus on finding carcasses following collision, estimating the number of carcasses missed through searcher efficiency, carcass persistence trials and carcass fall distributions, and modeling statistically robust fatality rates. Few technologies have been developed to monitor offshore bird and bat collisions, and many that have been developed focused on detecting collisions with large birds. The few studies that have attempted to document collisions at offshore turbines do not account for smaller bodied animals or for collisions that might be missed, which prevents the calculation of statistically robust fatality rates. The overall goal of this report, A Multi-Sensor Approach for Measuring Bird and Bat Collisions with Offshore Wind Turbines (Project), was to develop an effective multi-sensor system for quantifying bird and bat collision rates, specifically for offshore wind facilities. The Project goal and resulting automated collision detection system was achieved through two major technological advancements: 1) refining The Netherlands Organisation for Applied Scientific Research’s (TNO’s) existing WT-Bird® vibration sensing system, that had successfully detected large bird collisions during daytime, to allow for improved detection of smaller birds and bats during both daytime and nighttime hours and 2) improving image processing systems and developing and integrating machine learning algorithms to automatically detect and classify small and large bird and bat collisions with offshore turbines. This final technical report (FTR) summarizes Methods , Results , Conclusions , and Lessons Learned during each of the five Tasks identified for this research and development effort. This FTR includes summaries of the following: Task 1. Initial Engineering Tests to Improve WT-Bird® Task 2. Installation of WT‐Bird® on a Utility-scale Turbine at the National Wind Technology Center – National Renewable Energy Laboratory Task 3. Field Tests and Refinement of the Object Detection System Task 4. Validation of WT-Bird® on a Land-based Turbine Task 5. Preparation for the Implementation of WT-Bird® on an Offshore Turbine. This research and development effort documented successful improvement of the WT Bird® collision detection system to detect small birds and bats, and WT-Bird® is the first collision detection system to validate results compared to land-based post-construction monitoring. The collision trials provide estimates of missed targets that can be used to estimate fatality rates, a significant improvement relative to other offshore collision monitoring systems. Advances were made in developing an edge-processing solution to reduce data storage requirements, which is important if the system is deployed for long periods of time at offshore turbines. The improved WT-Bird® system also provides an important option for wind operators on land or offshore who need to document specific details about when collisions occur, particularly efforts to further research on bat impact minimization, or when standard fatality searches are impractical (e.g. offshore) or inadequate (e.g. challenging locations on land).

17 WIND ENERGY↗

Improving materials property predictions for graph neural networks with minimal feature engineering *

Graph neural networks (GNNs) have been employed in materials research to predict physical and functional properties, and have achieved superior performance in several application domains over prior machine learning approaches. Recent studies incorporate features of increasing complexity such as Gaussian radial functions, plane wave functions, and angular terms to augment the neural network models, with the expectation that these features are critical for achieving a high performance. Here, we propose a GNN that adopts edge convolution where hidden edge features evolve during training and extensive attention mechanisms, and operates on simple graphs with atoms as nodes and distances between them as edges. As a result, the same model can be used for very different tasks as no other domain-specific features are used. With a model that uses no feature engineering, we achieve performance comparable with state-of-the-art models with elaborate features for formation energy and band gap prediction with standard benchmarks; we achieve even better performance when the dataset size increases. Although some domain-specific datasets still require hand-crafted features to achieve state-of-the-art results, our selected architecture choices greatly reduce the need for elaborate feature engineering and still maintain predictive power in comparison.

42 ENGINEERING↗

Securing Smart Manufacturing: Detection of Cyber-Physical Attacks in CNC-Based Systems

As Industry 4.0 advances, the integration of computer numerical control (CNC) machines and advanced manufacturing technologies is transforming production into smart manufacturing systems that blend physical and digital processes as cyber-physical systems. However, this increased cyber-physical connectivity exposes manufacturing systems to cyber threats that can cause severe operational and financial disruptions. This paper presents a comparative study on cyber attacks and anomaly detection techniques in manufacturing, focusing on network traffic from CNC machines. The data extracted from network packets includes machine commands and control signals exchanged between the machine's interface and control system, crucial for maintaining operational integrity. We explore two types of cyber attacks, design modification and command injection, which pose substantial risks to CNC machine productivity and system integrity. Our investigation involves experiments on a real CNC system, highlighting the urgent need for effective detection mechanisms. To address these threats, we evaluate three anomaly detection methods: dynamic time warping (DTW), rolling average, and a deep learning, long short-term memory (LSTM) time-series-based autoencoder. Each is assessed for its effectiveness in identifying anomalous behaviors caused by the attacks. Our findings demonstrate the unique strengths and limitations of each detection technique, providing a deeper understanding of their applicability in realworld manufacturing environments. The comparative analysis indicates that while certain methods are highly effective against specific attack types, others offer broader applicability across different attacks. This study contributes to the accurate detection of anomalies in CNC machining processes, thereby enhancing the reliability and security of smart manufacturing systems against diverse cyber threats.

Williams, Bethanie [Tennessee Technological Univer↗

Machine Learning for Improved Availability of the SNS Klystron High Voltage Converter Modulators

Beam availability has increased at the SNS, however, the targeted availability is greater than 95 %, while the SNS has failed to meet lower targets in the past. The HVCM used to power the linac klystrons have been one source of lost beam time and was chosen to explore using AI/ML techniques to improve reliability. Among the possibilities being explored are automating the tuning of HVCMs and predicting component failures such as capacitor aging, rectifier assemblies containing hundreds of diodes, and insulating oil degradation. The methodology pursued includes data cleaning, de-noising, post-analysis data labeling, and machine learning model development. We explore using Long Short-Term Memory and autoencoders for anomaly detection and prognostication used to schedule maintenance. We evaluate the use of model regularizers and constraints to improve the performance of the model and investigate methods to estimate the uncertainty of the models to provide a robust prediction with statistical interoperability. This paper describes the operational experience and known failures of the HVCMs and the proposed ML methodology and the preliminary results of training the AI/ML algorithms.

Pappas, G. C.↗

Field Emission Mitigation in CEBAF SRF Cavities Using Deep Learning

The Continuous Electron Beam Accelerator Facility (CEBAF) operates hundreds of superconducting radio frequency (SRF) cavities in its two main linear accelerators. Field emission can occur when the cavities are set to high operating RF gradients and is an ongoing operational challenge. This is especially true in newer, higher gradient SRF cavities. Field emission results in damage to accelerator hardware, generates high levels of neutron and gamma radiation, and has deleterious effects on CEBAF operations. So, field emission reduction is imperative for the reliable, high gradient operation of CEBAF that is required by experimenters. Here we explore the use of deep learning architectures via multilayer perceptron to simultaneously model radiation measurements at multiple detectors in response to arbitrary gradient distributions. These models are trained on collected data and could be used to minimize the radiation production through gradient redistribution. This work builds on previous efforts in developing machine learning (ML) models, and is able to produce similar model performance as our previous ML model without requiring knowledge of the field emission onset for each cavity.

Ahammed, K.↗

Conduction-Cooled Operation of an SRF Multi-Cell Cavity

The development of compact, SRF-based accelerators for applications beyond research is experiencing notable advancements due to the use of cryocoolers for conduction-cooling instead of traditional liquid cryogens. Following the successful demonstration of a single-cell cavity operated through conduction cooling with three two-stage cryocoolers, Jefferson Lab (JLab) has made strides in the operation of a multi-cell resonator. This milestone paves the way for high-energy applications of compact, conduction-cooled SRF machines. The demonstration, carried out in collaboration with General Atomics, will take place in a dedicated horizontal test cryostat (HTC) at their San Diego facility. This presentation will highlight the technological developments, the latest results, and valuable lessons learned.

Vennekate, J. [Thomas Jefferson National Accelerat↗

Industrial Internet-of-Things & Data Analytics for Nuclear Power & Safeguards

Data analytics applied to nuclear power operations and nuclear safeguards is in a nascent state, yet some significant initial efforts are being undertaken by industry and academia. This report highlights our findings as to the current state-of-the-art of such efforts, in particular considering the Industrial Internet-of-Things aspect of this work, as well as an investigation into the utility of machine learning tools being developed for other industries. Blockchain applications were also studied. A consideration was undertaken into how to apply data analytics and machine learning to nuclear power and safeguards within the realm of Probabilistic Risk Assessments (PRAs), predictive maintenance & edge analytics, and proprietary data sharing.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗