Search NASA⌕ Search

SEARCH · Search NASA

Results for “tensorflow”

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.

108 records · Page 6

DELTA: An Open-Source Framework to Simplify Deep Learning with Satellite Imagery

DELTA (Deep Earth Learning, Tools, and Analysis) is an open-source framework developed at NASA for deep learning on satellite imagery based on tensorflow. It helps simplify data engineering and preprocessing steps and reduces the need for a lot of the boilerplate code that needs written to make datasets palatable for machine learning. This lets data scientists focus on model development while DELTA handles the grunt work. This presentation will demonstrate DELTA’s functionality and share some examples from an active project using it for flood mapping.

Michael von Pohle↗

Design of a Low-Cost, Submersible, Digital Holographic Microscope for in Situ Microbial Imaging

The methodologies for studying marine microbiology typically consist of utilizing instrumentation within a laboratory. This typically requires extracting a sample from its place of origin prior to examination, which may be days after collection. Oftentimes, the solution is to bring the lab to the ocean, which may be costly and provide further limitations for a sterile and stable laboratory environment. Here we present a low-cost, submersible, digital holographic microscope (DHM) designed to image marine microorganisms (such as bacteria and plankton) in their natural underwater environment. Our instrument eliminates the need to transport samples and allows for instantaneous data collection of microbes in-situ. The DHM achieves sub-micron spatial resolution and is paired with artificial intelligence for the detection and tracking of specimens to reduce the overall collected data. This instrument also aims to reduce the cost of manufacturing and field expenses relative to marine microbiological research. “Off the shelf” components were selected in the design process of this instrument which allows us to achieve precise results without sacrificing data quality. The DHM itself costs under one thousand dollars and features a low-cost high-resolution camera, the Arducam MT9J001. Included in the design were five main subsystems: optical, mechanical, electrical, power, and machine learning. Our on board computer and artificial intelligence consist of a Raspberry Pi 4 (8 Gb) and Google Coral USB Tensorflow accelerator. Instrument testing has successfully proven our abilities of data acquisition for at least two hours in depths of at least forty meters below sea level. Additionally, our artificial intelligence system is currently capable of tracking up to ten areas of interest in a fraction of a second with over ninety percent confidence via the neural net driven by the tensor cores on the Google Coral. Furthermore, we have demonstrated the versatility of our instrument by mounting it on an ocean-going ROV, the BlueRov2 by BlueRobotics. Our tests on the BlueRov2 exemplified the cost-effective nature of a submersible and reusable instrument that can be implemented in moderate environments and on most vessels.

Wallace, James Kent↗

AMD Radeon e9173 Low Power PCIE GPU Single Event Effects Test Report

The AMD Radeon Embedded e9170 Graphics Processing Unit (GPU), notably the e9173 Peripheral Component Interconnect Express (PCIE) variant, is of interest to Artemis generation programs with requirements for graphics rendering, compute, artificial intelligence (Ai) with a constraints-requiring piece-part procurement and power consumption of less than 50W. In addition to collecting heavy ion data on this device, a secondary purpose of this test campaign was to validate video capture hardware and software workflows used with GPU, microprocessor and system-on-chip device testing. Five (5) test patterns from the NEPP Processor Enclave (NPE) test suite were used with the e9173. The test patterns covered the operating system’s (OS) idle contribution towards the cross section, matrix math using tensorflow-rocm, two artificial intelligence models developed at NASA GSFC, and an industry standard GPU benchmarking application called Mesa GLXGears.

NASA Technical Memorandum (TM) test report for pos↗

Deep Modeling: Circuit Characterization Using Theory Based Models in a Data Driven Framework

Analog computational circuits have been demonstrated to provide substantial improvements in power and speed relative to digital circuits, especially for applications requiring extreme parallelism but only modest precision. Deep machine learning is one such area and stands to benefit greatly from analog and mixed-signal implementations. However, even at modest precisions, offsets and non-linearity can degrade system performance. Furthermore, in all but the simplest systems, it is impossible to directly measure the intermediate outputs of all sub-circuits. The result is that circuit designers are unable to accurately evaluate the non-idealities of computational circuits in-situ and are therefore unable to fully utilize measurement results to improve future designs. In this paper we present a technique to use deep learning frameworks to model physical systems. Recently developed libraries like TensorFlow make it possible to use back propagation to learn parameters in the context of modeling circuit behavior. Offsets and scaling errors can be discovered even for sub-circuits that are deeply embedded in a computational system and not directly observable. The learned parameters can be used to refine simulation methods or to identify appropriate compensation strategies. We demonstrate the framework using a mixed-signal convolution operator as an example circuit.

Bolme, David S↗

Solving differential equations with unknown constitutive relations as recurrent neural networks

We solve a system of ordinary differential equations with an unknown functional form of a sink (reaction rate) term. We assume that the measurements (time series) of state variables are partially available, and use a recurrent neural network to “learn” the reaction rate from this data. This is achieved by including discretized ordinary differential equations as part of a recurrent neural network training problem. We extend TensorFlow’s recurrent neural network architecture to create a simple but scalable and effective solver for the unknown functions, and apply it to a fedbatch bioreactor simulation problem. Use of techniques from recent deep learning literature enables training of functions with behavior manifesting over thousands of time steps. Our networks are structurally similar to recurrent neural networks, but differ in purpose, and require modified training strategies.

deep learning↗

Exploring flexible communications for streamlining DNN ensemble training pipelines

Parallel training of a Deep Neural Network (DNN) ensemble on a cluster of nodes is a common practice to train multiple models in order to construct a model with a higher prediction accuracy, or to quickly tune the parameters of a training model. Existing ensemble training pipelines perform a great deal of redundant operations, resulting in unnecessary CPU usage, or even poor pipeline performance. In order to remove these redundancies, we need pipelines with more communication flexibility than existing DNN frameworks can provide. This project investigates a series of designs to improve pipeline flexibility and adaptivity, while also increasing performance. We implement our designs using Tensorflow with Horovod, and test it using several large DNNs in a large scale GPU cluster, the Titan supercomputer at Oak Ridge National Lab. Our results show that with the new flexible communication schemes, the CPU time spent during training is reduced by 2-11X. Furthermore, our implementation can achieve up to 10X speedups when CPU core limits are imposed. Our best pipeline also reduces the average power draw of the ensemble training process by 5--16% when compared to the baseline.

Pittman, Randall↗

Fishing for Clickbaits in Social Images and Texts with Linguistically-Infused Neural Network Models

This paper presents the results and conclusions of our participation in the Clickbait Challenge 2017 on automatic clickbait detection in social media. We first describe linguistically-infused neural network models and identify informative representations to predict the level of clickbaiting present in Twitter posts. Our models allow to answer the question not only whether a post is a clickbait or not, but to what extent it is a clickbait post e.g., not at all, slightly, considerably, or heavily clickbaity using a score ranging from 0 to 1. We evaluate the predictive power of models trained on varied text and image representations extracted from tweets. Our best performing model that relies on the tweet text and linguistic markers of biased language extracted from the tweet and the corresponding page yields mean squared error (MSE) of 0.04, mean absolute error (MAE) of 0.16 and R2 of 0.43 on the held-out test data. For the binary classification setup (clickbait vs. non-clickbait), our model achieved F1 score of 0.69. We have not found that image representations combined with text yield significant performance improvement yet. Nevertheless, this work is the first to present preliminary analysis of objects extracted using Google Tensorflow object detection API from images in clickbait vs. non-clickbait Twitter posts. Finally, we outline several steps to improve model performance as a part of the future work.

dlickbait challenge, Twitter, neural networks, mac↗

Subsurface Characterization and Machine Learning Predictions at Brady Hot Springs: Preprint

Subsurface data analysis, reservoir modeling, and machine learning (ML) techniques have been applied to the Brady Hot Springs (BHS) geothermal field in Nevada, USA to further characterize the subsurface and assist with optimizing reservoir management. Hundreds of reservoir simulations have been conducted in TETRAD-G and CMG STARS to explore different injection and production fluid flow rates and allocations and to develop a training data set for ML. This process included simulating the historical injection and production since 1979 and prediction of future performance through 2040. ML networks were created and trained using TensorFlow based on multilayer perceptron (MLP), long short-term memory (LSTM), and convolutional neural network (CNN) architectures. These networks took as input selected flow rates, injection temperatures, and historical field operation data and produced estimates of future production temperatures. This approach was first successfully tested on a simplified single fracture doublet system, followed by the application to the BHS reservoir. Using an initial BHS dataset with 37 simulated scenarios, the trained and validated network predicted the production temperature for 6 production wells with the mean absolute percentage error of less than 8%. In a complementary analysis effort, the principal component analysis applied to 13 BHS geological parameters revealed that vertical fracture permeability shows the strongest correlation with fault density and fault intersection density. A new BHS reservoir model was developed considering the fault intersection density as proxy for permeability. This new reservoir model helps to explore under-exploited zones in the reservoir. A data gathering plan to obtain additional subsurface data was developed; it includes temperature surveying for three idle injection wells, at which the reservoir simulations indicate high bottom-hole temperatures. The collected data assist with calibrating the reservoir model and may lead to converting these wells to producers to access under-exploited zones in the reservoir. Data gathering activities are planned for the first quarter of 2021.

40 EE - Geothermal Technologies Office (EE-4G)↗

Deep learning for lipid droplet recognition in quantitative phase images

This library of Python code is used for performing semantic segmentation of images using 6 different machine learning methods. Five of the methods are implemented entirely within the scikit-learn framework. The Convolutional Neural Network (CNN) method requires Keras with a TensorFlow backend and generally uses a different set of scripts in order to perform the complete training and evaluation.

Sheneman, Lucas↗

Autonomous Navigation and Control of UGVs' in Nuclear Power Plants - 20381

The purpose of the husky A200 ground robot is to autonomously navigate through the places where it is very hazardous for human beings to reach and operate, like nuclear power plants, chemical industries. The aim is to navigate the ground robot autonomously with an Arm mounted on the robot along with the different sensors as camera, and Lidar. The autonomous motion of the robot is controlled by the controller which uses path planner for trajectory generation of the robot. The mission planner uses the current position of the husky A200, given the way points of the initial and the destination it would extract a best possible route based on the current events provided using GMapping. The global reference frame is used for planning the way points. Creating the appropriate path and the actions required to follow the path are given by the motion planner. The motion planner depends on the active sensor data such as obstacles, lanes, based on the sensor data feasible path is generated. Feasibility of the path is determined by the dynamics of the husky and a series of points generated with certain velocity and acceleration profile. The controller adjusts the lateral, longitudinal and yaw motion of the husky to command the behaviors. The kinematic model is developed for kinematic motion of the husky and the dynamic model is developed for transient and steady state characteristics. The images and other type of data captured by the camera are processed through the computational framework used to build machine learning models. TensorFlow will be used for deep learning and to identify and classify different objects around the husky. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Development of neural network force fields for corrosion studies

To fully understand the chemistry and physics of corrosion, novel methods of simulation must be developed. One approach is designing machine learning (ML) algorithms integrated with density functional theory to develop adaptive force fields to gain insight into corrosion behavior namely at the surface of metal oxides. Current methods of modeling corrosion are slow due to the computational cost of resolving both reaction mechanics and mass transport processes. Machine learning methods can be implemented to obtain structure-activity relationships at both the molecular and bulk scale while still retaining the accuracy of density functional theory (DFT) and significantly decreasing the time needed for simulations of complex chemical processes in the various environments of corrosion. Multiscale models are needed for corrosion studies to fully understand its processes not only at the atomic length scale (chemical bonding, energies, and forces), but also at the nano and meso length scales (solid-state physics and material science processes). Current methods of study include DFT, molecular dynamics, and Monte Carlo. The limitation of DFT is that only a small number of atoms or molecules can be simulated at that level of theory. Density functional theory is used to study the electronic structure of atoms and molecules, and calculate the force component of each atom. However, these calculations are limited to about 1000 atoms. Custom periodic boundary conditions (PBC) can be used to describe the various environments and defects that affect the atomic forces to produce a large data set from which a training set can be derived. Machine learning can be utilized to overcome the barrier of modeling macroscopic and multi-scale processes from ab initio calculations through the development of adaptive force fields. Local environments determine the atomic forces of a given system, therefore adaptive force fields must be created to produce reliable quantum mechanical calculations. This can be achieved by developing a learning algorithm that uses the mapped atomic forces or fingerprint as an input to produce energies and magnetic moments as output. A systematic approach was used to begin to build a data set in order to accurately describe the atomic forces in various environments. In Figure 4 below, a simple PBC cell of Fe{sub 2}O{sub 3} was first optimized. A surface optimization was performed next, followed by a hydroxylated surface optimization. Once this calculation has converged, the adsorption of halide species to the hydroxylated surface will be investigated. TensorFlow is an open source platform for machine learning developed by Google. Using a high level application program interface (API) such as Keras allows for building and training ML models easily in a number of different environments and languages. For this project, a neural network was developed within Anaconda in Python. Future Work: Further development of reference data set; Refining neural network and learning algorithm; Fingerprinting atomic environment to enable mapping of atomic force components; Choosing appropriate training set from reference data; Learning from training set and enabling non-linear mapping of training set fingerprints and the atomic forces; Estimation of uncertainty to identify ranges of outside applicability; Testing and analysis of molecular dynamic simulations.

36 MATERIALS SCIENCE↗

Programs and Code for Geothermal Exploration Artificial Intelligence

The scripts below are used to run the Geothermal Exploration Artificial Intelligence developed within the "Detection of Potential Geothermal Exploration Sites from Hyperspectral Images via Deep Learning" project. It includes all scripts for pre-processing and processing, including: - Land Surface Temperature K-Means classifier - Labeling AI using Self Organizing Maps (SOM) - Post-processing for Permanent Scatterer InSAR (PSInSAR) analysis with SOM - Mineral marker summarizing - Artificial Intelligence (AI) Data splitting: creates data set from a single raster file - Artificial Intelligence Model: creates AI from a single data set, after splitting in Train, Validation and Test subsets - AI Mapper: creates a classification map based on a raster file

15 GEOTHERMAL ENERGY↗

Subsurface Characterization and Machine Learning Predictions at Brady Hot Springs Results

Geothermal power plants typically show decreasing heat and power production rates over time. Mitigation strategies include optimizing the management of existing wells - increasing or decreasing the fluid flow rates across the wells - and drilling new wells at appropriate locations. The latter is expensive, time-consuming, and subject to many engineering constraints, but the former is a viable mechanism for periodic adjustment of the available fluid allocations. Data and supporting literature from a study describing a new approach combining reservoir modeling and machine learning to produce models that enable strategies for the mitigation of decreased heat and power production rates over time for geothermal power plants. The computational approach used enables translation of sets of potential flow rates for the active wells into reservoir-wide estimates of produced energy and discovery of optimal flow allocations among the studied sets. In our computational experiments, we utilize collections of simulations for a specific reservoir (which capture subsurface characterization and realize history matching) along with machine learning models that predict temperature and pressure timeseries for production wells. We evaluate this approach using an "open-source" reservoir we have constructed that captures many of the characteristics of Brady Hot Springs, a commercially operational geothermal field in Nevada, USA. Selected results from a reservoir model of Brady Hot Springs itself are presented to show successful application to an existing system. In both cases, energy predictions prove to be highly accurate: all observed prediction errors do not exceed 3.68% for temperatures and 4.75% for pressures. In a cumulative energy estimation, we observe prediction errors that are less than 4.04%. A typical reservoir simulation for Brady Hot Springs completes in approximately 4 hours, whereas our machine learning models yield accurate 20-year predictions for temperatures, pressures, and produced energy in 0.9 seconds. This paper aims to demonstrate how the models and techniques from our study can be applied to achieve rapid exploration of controlled parameters and optimization of other geothermal reservoirs. Includes a synthetic, yet realistic, model of a geothermal reservoir, referred to as open-source reservoir (OSR). OSR is a 10-well (4 injection wells and 6 production wells) system that resembles Brady Hot Springs (a commercially operational geothermal field in Nevada, USA) at a high level but has a number of sufficiently modified characteristics (which renders any possible similarity between specific characteristics like temperatures and pressures as purely random). We study OSR through CMG simulations with a wide range of flow allocation scenarios. Includes a dataset with 101 simulated scenarios that cover the period of time between 2020 and 2040 and a link to the published paper about this project, where we focus on the Machine Learning work for predicting OSR's energy production based on the simulation data, as well as a link to the GitHub repository where we have published the code we have developed (please refer to the repository's readme file to see instructions on how to run the code). Additional links are included to associated work led by the USGS to identify geologic factors associated with well productivity in geothermal fields. Below are the high-level steps for applying the same modeling + ML process to other geothermal reservoirs: 1. Develop a geologic model of the geothermal field. The location of faults, upflow zones, aquifers, etc. need to be accounted for as accurately as possible 2. The geologic model needs to be converted to a reservoir model that can be used in a reservoir simulator, such as, for instance, CMG STARS, TETRAD, or FALCON 3. Using native state modeling, the initial temperature and pressure distributions are evaluated, and they become the initial conditions for dynamic reservoir simulations 4....

15 GEOTHERMAL ENERGY↗

Battery inverter experimental data

The increase in power electronic based generation sources require accurate modeling of inverters. Accurate modeling requires experimental data over wider operation range. We used 30 kW off-the-shelf grid following battery inverter in the experiments. We used controllable AC supply and controllable DC supply to emulate AC and DC side characteristics. The experiments were performed at NREL's Energy Systems Integration Facility. Inverter is tested under 100%, 75%, 50%, 25% load conditions. In the first dataset, for each operating condition, controllable AC source voltage is varied from 0.9 to 1.1 per unit (p.u) with a step value of 0.025 p.u while keeping the frequency at 60 Hz. In the second dataset, under similar load conditions (100%, 75%, 50%, 25% ), the frequency of the controllable AC source voltage was varied from 59 Hz to 61 Hz with a step value of 0.2 Hz. Voltage and frequency range is chosen based on inverter protection. Voltages and currents on DC and AC side are included in the dataset.

24 POWER TRANSMISSION AND DISTRIBUTION↗

PV inverter experimental data

The increase in power electronic based generation sources require accurate modeling of inverters. Accurate modeling requires experimental data over wider operation range. We used 20 kW off-the-shelf grid following PV inverter in the experiments. We used controllable AC supply and controllable DC supply to emulate AC and DC side characteristics. The experiments were performed at NREL's Energy Systems Integration Facility. Due to the limitations of the DC supply used, inverter is tested under 75%, 50%, 25% load conditions (This dataset does not contain 100% load condition). In the first dataset, for each operating condition, controllable AC source voltage is varied from 0.88 to 1.09 per unit (p.u) with a step value of 0.025 p.u while keeping the frequency at 60 Hz. In the second dataset, under similar load conditions (75%, 50%, 25% ), the frequency of the controllable AC source voltage was varied from 59.4 Hz to 60.45 Hz with a step value of 0.2 Hz. Voltage and frequency range is chosen based on inverter protection. Voltages and currents on DC and AC side are included in the dataset.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Split phase inverter data

The increase in power electronic based generation sources require accurate modeling of inverters. Accurate modeling requires experimental data over wider operation range. We used 8.35 kW off-the-shelf grid following split phase PV inverter in the experiments. We used controllable AC supply and controllable DC supply to emulate AC and DC side characteristics. The experiments were performed at NREL's Energy Systems Integration Facility. Inverter is tested under 100%, 75%, 50%, 25% load conditions. In the first dataset, for each operating condition, controllable AC source voltage is varied from 0.9 to 1.1 per unit (p.u) with a step value of 0.025 p.u while keeping the frequency at 60 Hz. In the second dataset, under similar load conditions (100%, 75%, 50%, 25% ), the frequency of the controllable AC source voltage was varied from 59 Hz to 61 Hz with a step value of 0.2 Hz. Voltage and frequency range is chosen based on inverter protection. Voltages and currents on DC and AC side are included in the dataset.

24 POWER TRANSMISSION AND DISTRIBUTION↗

PV Inverter Experimental Dataset Version 2 with 100 Percent Power

The increase in power electronic based generation sources require accurate modeling of inverters. Accurate modeling requires experimental data over wider operation range. We used 20 kW off-the-shelf grid following PV inverter in the experiments. We used controllable AC supply and controllable DC supply to emulate AC and DC side characteristics. The experiments were performed at NREL's Energy Systems Integration Facility. The PV inverter is tested under 100%, 75%, 50%, 25% load conditions. In the first dataset, for each operating condition, controllable AC source voltage is varied from 0.88 to 1.09 per unit (p.u) with a step value of 0.025 p.u while keeping the frequency at 60 Hz. In the second dataset, under similar load conditions (100%, 75%, 50%, 25% ), the frequency of the controllable AC source voltage was varied from 59.4 Hz to 60.45 Hz with a step value of 0.2 Hz. Voltage and frequency range is chosen based on inverter protection. Voltages and currents on DC and AC side are included in the dataset.

14 SOLAR ENERGY↗