ML UT NDE prediction for FSP metal

This is a repository supporting the paper 'ML-Statistical Ultrasonic Non-Destructive Prediction for Friction Stir Processed Metal'

This code is provided to reproduce the results and figures presented throughout the manuscript which investigated ML methods for UT NDE prediction for FSP metal.

Getting started

1. Clone the repository and navigate to the project folder.
2. Download four data folders and make note of the local paths to these data folders.
   • FSW Process Data: https://data.pnnl.gov/group/nodes/dataset/34283
   • MCPC Round 1: https://data.pnnl.gov/group/nodes/dataset/34281
   • MCPC Round 2: https://data.pnl.gov/group/65/nodes/dataset/36380
   • MCPC Round 3: https://data.pnl.gov/group/65/nodes/dataset/36575
3. Run the code sequentially starting with the scripts in code/00_wrangle/00_collect_paths, followed by the scripts in code/00_wrangle/01_filter_wrangle, and so on.
   • Local paths to the downloaded data folders will be required for the scripts in code/00_wrangle/00_collect_paths.
   • Packages required by library() commands in script preambles should be installed using install.packages().
4. Resultant wrangled data, model results, and figures should populate in top-level data and results folders.

Top level folders

• code: contains three folders that should be run sequentially:
  • 00_wrangle: wrangle and join the large MCPC data from datahub into appropriate file formats for modeling.
  • 01_model: model the UT NDE signal using tidymodels workflow.
  • 02_grain_size: extend predictions to grain size prediction.
  • 03_sensitivity_analysis: sensitivity analysis testing the stability of random forest rankings.
• data: contains small manually curated files needed for wrangling the data and extending to grain size prediction.
• freeze_packages: see Notes on reproducibility below.
• results (untracked): must be created, will store various plots and output objects throughout the modeling pipeline.

Notes on reproducibility

All package versions (R and python) were frozen after the revision and can be reviewed in the freeze_packages folder. For details on reproducibility, please see the supplementary material associated with the manuscript.

Disclaimer

This material was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the United States Department of Energy, nor Battelle, nor any of their employees, nor any jurisdiction or organization that has cooperated in the development of these materials, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness or any information, apparatus, product, software, or process disclosed, or represents that its use would not infringe privately owned rights.

Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or Battelle Memorial Institute. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

             PACIFIC NORTHWEST NATIONAL LABORATORY
                          operated by
                            BATTELLE
                            for the
               UNITED STATES DEPARTMENT OF ENERGY
                under Contract DE-AC05-76RL01830