/A-machine-learning-pipeline-for-estimation-of-Lithium-ion-battery-state-of-health

Primary LanguagePython

A-machine-learning-pipeline-for-estimation-of-Lithium-ion-battery-state-of-health

Data used in the analysis:

• Group 1: https://web.calce.umd.edu/batteries/data.htm https://ti.arc.nasa.gov/tech/dash/groups/pcoe/prognostic-data-repository/

• Group 2: https://data.matr.io/1/projects/5c48dd2bc625d700019f3204

• Group 3: https://ora.ox.ac.uk/objects/uuid:03ba4b01-cfed-46d3-9b1a-7d4a7bdf6fac

Python packages required to run code (any version is applicable) for dependencies refer to developer website:

  • pandas
  • numpy
  • scikit-learn
  • matplotlib
  • tensorflow

How to run code:

  1. Download supplied data. Note: Data is for demo code only and does not include all datasets used in the paper. Data name indicates what the data should be used for.
  2. Load data and run feature_selection.py
  3. Load saved data in any of the supplied model code for training
  4. Load the trained model in the supplied jupyter notebook results.ipynb

NOTE: For access to the data processing and feature engineering code, please contact Darius Roman at dvr1@hw.ac.uk for the academic license. Only the data modelling code is available without agreeing to a license.