Skip to content

README.md

Latest commit

 

History

History
109 lines (87 loc) · 3.28 KB

README.md

File metadata and controls

109 lines (87 loc) · 3.28 KB

AITW stiffness surrogate model

NOTES Requires python3.10 for the current pinned version of tensorflow.

Installaton

OPTIONAL Create and initialize a clean virtual environment

python3.10 -m venv "surrogate-virtual-env"
source surrogate-virtual-env/bin/activate

Install dependencies

cd <PATH to folder with pyproject.toml>
pip install .

Usage

CLI

The code makes avaialble a CLI utility to run the inference given an input JSON file.

aitw-stiffness example-input/input.json output.json 0.05 0.95

Adjust the parameters as needed (Use --help for more information).

Parameters

Parameter Description Type
<input-path> File path to the input file containing a properly formatted JSON file. See fromat below. String
<output-path> File path for the stiffness estimate JSON file. Example output shown below. String
<quantile-lower-bound> Lower bound percentage of the uncertainty estimate. Float [ 0, 1]
<quantile-upper-bound> Upper bound percentage of the uncertainty estimate Float [ 0, 1]

Input parameter values and ranges

The surrogate has been trained with physical parameters ranging between the values below. Fiber orientation distribution parameter takes one of the two values below.
An error will be thrown if the input parameter values do not coincide with the limits.

Parameter Possible value
fiberOrientationDistribution "vonMises", "vonMisesFisher"
Parameter Minimum Maximum
fiberAspectRatio 1.2 20000
orientationConcentration 1e-10 100
microfibrilAngle 0 45
lumenPorosity 0 0.99
fiberVolumeFraction 0 1
celluloseContent 0.05 0.9
hemicelluloseContent 0.01 0.5
ligninContent 0.01 0.6
pectinContent 0 0.15
extractivesContent 0 0.15
ashContent 0 0.2
celluloseCrystallinity 0.2 1.0
matrixYoungsModulus 0 100
matrixPoissonRatio 0 0.5
airPorosity 0 0.5
tangentialInterfaceCompliance 1e-10 100000.0
longitudinalInterfaceCompliance 1e-10 100000.0

Output file example

This results in the following output file format in "./output.json":

{
  "E_T": {
    "median": 3.550992727279663,
    "lowerBound": 3.2585065364837646,
    "upperBound": 3.869732141494751
  },
  "E_L": {
    "median": 1.9295220375061035,
    "lowerBound": 1.6106904745101929,
    "upperBound": 2.3114655017852783
  },
  "nu_T": {
    "median": 0.2790219783782959,
    "lowerBound": 0.19306382536888123,
    "upperBound": 0.36498013138771057
  },
  "nu_LT": {
    "median": 0.19985055923461914,
    "lowerBound": 0.15456369519233704,
    "upperBound": 0.24513742327690125
  },
  "mu_LT": {
    "median": 1.482556939125061,
    "lowerBound": 1.3604427576065063,
    "upperBound": 1.61563241481781
  },
  "quantiles": [
    0.05,
    0.95
  ]
}