View: |
Part 1: Document Description
|
Citation |
|
---|---|
Title: |
Input-Output Dataset for Physics-inspired Artificial Neural Network for Dynamic System |
Identification Number: |
doi:10.18419/darus-633 |
Distributor: |
DaRUS |
Date of Distribution: |
2020-07-21 |
Version: |
1 |
Bibliographic Citation: |
Praditia, Timothy, 2020, "Input-Output Dataset for Physics-inspired Artificial Neural Network for Dynamic System", https://doi.org/10.18419/darus-633, DaRUS, V1 |
Citation |
|
Title: |
Input-Output Dataset for Physics-inspired Artificial Neural Network for Dynamic System |
Identification Number: |
doi:10.18419/darus-633 |
Authoring Entity: |
Praditia, Timothy (Universität Stuttgart) |
Other identifications and acknowledgements: |
Walser, Thilo |
Other identifications and acknowledgements: |
Oladyshkin, Sergey |
Other identifications and acknowledgements: |
Nowak, Wolfgang |
Grant Number: |
EXC-2075 – 390740016 |
Distributor: |
DaRUS |
Access Authority: |
Nowak, Wolfgang |
Depositor: |
Praditia, Timothy |
Date of Deposit: |
2020-02-07 |
Holdings Information: |
https://doi.org/10.18419/darus-633 |
Study Scope |
|
Keywords: |
Computer and Information Science, Earth and Environmental Sciences, Thermochemical Energy Storage, Porous medium flow |
Abstract: |
This dataset contains two .mat files, one pre-processed (direct simulation results) and the other one is with added noise. The simulated problem is a thermochemical energy storage problem using CaO/Ca(OH)2 as the material choice. This dataset is used as input-output data pairs necessary for training, validating, and testing the ANN. The input data consist of CaO density, Ca(OH)2 density, CaO specific heat capacity, Ca(OH)2, porosity, permeability, reaction rate constant, initial and outlet pressure, initial temperature, inlet temperature, N2 molar inflow rate, H2O molar inflow rate, and specific reaction enthalpy. The output data consist of pressure, temperature, CaO volume fraction, and H2O molar fraction. Additionally, there is an automated script file for the DuMuX run. |
Methodology and Processing |
|
Sources Statement |
|
Data Access |
|
Other Study Description Materials |
|
Related Studies |
|
Praditia, T. (2020): Trained ANN Parameters for Physics-inspired Artificial Neural Network for Dynamic System, <a href="https://doi.org/10.18419/darus-634">doi: 10.18419/darus-634</a>, DaRUS. |
|
Related Publications |
|
Citation |
|
Title: |
Praditia, T., Walser, T., Oladyshkin, S. and Nowak, W.: Improving Thermochemical Energy Storage dynamics forecast with Physics-Inspired Neural Network architecture. Energies 2020 |
Bibliographic Citation: |
Praditia, T., Walser, T., Oladyshkin, S. and Nowak, W.: Improving Thermochemical Energy Storage dynamics forecast with Physics-Inspired Neural Network architecture. Energies 2020 |
Label: |
io_data.mat |
Text: |
This .mat file contains unprocessed input-output data pairs needed to train, validate, and test the ANN. |
Notes: |
application/matlab-mat |
Label: |
noisy_data.mat |
Text: |
This .mat file contains processed input-output data pairs needed to train, validate, and test the ANN. |
Notes: |
application/matlab-mat |
Label: |
README.md |
Text: |
README file explaining how the data is formatted |
Notes: |
text/markdown |
Label: |
run_script.sh |
Text: |
An automated script to run all samples in DuMuX for reproduction of the input-output data contained in the 'io_data.mat' and 'noisy_data.mat'. Before running, the path to the DuMuX executables needs to be adjusted accordingly. |
Notes: |
text/x-sh |