Processed Lab Data for Neural Network-Based Shear Stress Level Prediction

Cheng, Duo

doi:10.15121/1787545

Title: Processed Lab Data for Neural Network-Based Shear Stress Level Prediction

Dataset
Other Related Research

Abstract

Machine learning can be used to predict fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. The files are extracted features and labels from lab data (experiment p4679). The features are extracted with a non-overlapping window from the original acoustic data. The first column is the time of the window. The second and third columns are the mean and the variance of the acoustic data in this window, respectively. The 4th-11th column is the the power spectrum density ranging from low to high frequency. And the last column is the corresponding label (shear stress level). The name of the file means which driving velocity the sequence is generated from. Data were generated from laboratory friction experiments conducted with a biaxial shear apparatus. Experiments were conducted in the double direct shear configuration in which two fault zones are sheared between three rigid forcing blocks. Our samples consisted of two 5-mm-thick layers of simulated fault gouge with a nominal contact area of 10 by 10 cm^2. Gouge material consisted of soda-lime glass beads with initial particle size between 105 and 149 micrometers. Prior to shearing, we impose a constant fault normal stressmore »« less

Authors:

Cheng, Duo

Pennsylvania State University

Publication Date:: Fri May 14 00:00:00 EDT 2021

Other Number(s):: 1312

DOE Contract Number:: EE0008763

Research Org.:: DOE Geothermal Data Repository; Pennsylvania State University

Sponsoring Org.:: USDOE Office of Energy Efficiency and Renewable Energy (EERE), Renewable Power Office. Geothermal Technologies Program (EE-4G)

Collaborations:: Pennsylvania State University

Subject:: 15 GEOTHERMAL ENERGY; Biaxial Shear Experiment; MATLAB; acoustic emissions; acoustics; ai; artificial intelligence; biaxial shear apparatus; code; deep learning; energy; experiment; experimental data; fault; fault properties; friction; geophysics; geothermal; lab data; machine learning; microseismicity; processed data; seismic; seismic forcasting; seismic prediction; shear stress; time to failure

OSTI Identifier:: 1787545

DOI:: https://doi.org/10.15121/1787545

Citation Formats


                    Cheng, Duo. Processed Lab Data for Neural Network-Based Shear Stress Level Prediction.  United States: N. p., 2021. 
        Web.  doi:10.15121/1787545.

Copy to clipboard


                    Cheng, Duo. Processed Lab Data for Neural Network-Based Shear Stress Level Prediction.  United States.  doi:https://doi.org/10.15121/1787545

Copy to clipboard


                    Cheng, Duo. 2021.  
"Processed Lab Data for Neural Network-Based Shear Stress Level Prediction".  United States.  doi:https://doi.org/10.15121/1787545.  https://www.osti.gov/servlets/purl/1787545. Pub date:Fri May 14 00:00:00 EDT 2021

Copy to clipboard


                    
@article{osti_1787545,

  title        = {Processed Lab Data for Neural Network-Based Shear Stress Level Prediction},

  author       = {Cheng, Duo},

  abstractNote = {Machine learning can be used to predict fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. The files are extracted features and labels from lab data (experiment p4679). The features are extracted with a non-overlapping window from the original acoustic data. The first column is the time of the window. The second and third columns are the mean and the variance of the acoustic data in this window, respectively. The 4th-11th column is the the power spectrum density ranging from low to high frequency. And the last column is the corresponding label (shear stress level). The name of the file means which driving velocity the sequence is generated from. Data were generated from laboratory friction experiments conducted with a biaxial shear apparatus. Experiments were conducted in the double direct shear configuration in which two fault zones are sheared between three rigid forcing blocks. Our samples consisted of two 5-mm-thick layers of simulated fault gouge with a nominal contact area of 10 by 10 cm^2. Gouge material consisted of soda-lime glass beads with initial particle size between 105 and 149 micrometers. Prior to shearing, we impose a constant fault normal stress of 2 MPa using a servo-controlled load-feedback mechanism and allow the sample to compact. Once the sample has reached a constant layer thickness, the central block is driven down at constant rate of 10 micrometers per second. In tandem, we collect an AE signal continuously at 4 MHz from a piezoceramic sensor embedded in a steel forcing block about 22 mm from the gouge layer The data from this experiment can be used with the deep learning algorithm to train it for future fault property prediction.},

  doi          = {10.15121/1787545},

  journal      = {},

  number       = ,

  volume       = ,

  place        = {United States},

  year         = {Fri May 14 00:00:00 EDT 2021},

  month        = {Fri May 14 00:00:00 EDT 2021}

}

Copy to clipboard

Dataset:

View Dataset

DOI: https://doi.org/10.15121/1787545

Save / Share:

Export Metadata

Save to My Library

Similar records in DOE Data Explorer and OSTI.GOV collections:

Data from: Learning coagulation processes with combinatorially-invariant neural networks

Dataset Wang, Justin ; Curtis, Jeffrey H ; Riemer, Nicole ; ...

This dataset contains all the necessary information to recreate the study presented in the paper entitled "Learning coagulation processes with combinatorially-invariant neural networks". This consists of (1) the aggregated output files used for machine learning, (2) the machine learning codes used to learn the presented models, (3) the PartMC model source code that was used to generate the simulation data and (4) the Python scripts used construct the scenario library for training and testing simulations. This data was used to investigate a method (combinatorally-invariant neural network) for learning the aerosol process of coagulation. This data may be useful for applicationmore » of other methods.« less
Prediction of electron density and pressure profile shapes on NSTX-U using neural networks

Dataset Boyer, Mark ; Chadwick, Jason

A new model for prediction of electron density and pressure profile shapes on NSTX and NSTX-U has been developed using neural networks. The model has been trained and tested on measured profiles from experimental discharges during the first operational campaign of NSTX-U. By projecting profiles onto empirically derived basis functions, the model is able to efficiently and accurately reproduce profile shapes. In order to project the performance of the model to upcoming NSTX-U operations, a large database of profiles from the operation of NSTX is used to test performance as a function of available data. The rapid execution time ofmore »« less
A convolutional neural network-based screening tool for X-ray serial crystallography

Dataset Ke, Tsung-Wei
Data for: "Data efficiency and extrapolation trends in neural network interatomic potentials"

Dataset Schwalbe Koda, Daniel ; Vita, Joshua

Data to reproduce the paper "Data efficiency and extrapolation trends in neural network interatomic potentials" (LLNL-JRNL-845001). Contains: plots, results, and visualization scripts. Preprint available at https://doi.org/10.48550/arXiv.2302.05823.
Deep convolutional neural networks for multi-scale time-series classification and application to disruption prediction in fusion devices

Dataset Churchill, R M

The multi-scale, mutli-physics nature of fusion plasmas makes predicting plasma events challenging. Recent advances in deep convolutional neural network architectures (CNN) utilizing dilated convolutions enable accurate predictions on sequences which have long-range, multi-scale characteristics, such as the time-series generated by diagnostic instruments observing fusion plasmas. Here we apply this neural network architecture to the popular problem of disruption prediction in fusion tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. ECEi measures a fundamental plasma quantity (electron temperature) with high temporal resolution over the entire plasma discharge, making it sensitive to a number of potential pre-disruptions markers with different temporal and spatial scales. Promising, initial disruption prediction results are obtained training a deep CNN with large receptive field ({more »« less

Similar Records