Skip to content

Qulxis/ML-Climate-Project-ahx2001

 
 

Repository files navigation

ML-Climate-Project-Fall2022

Solar Panel Location Selection Using Power Generation Potential Prediction

This project is for my semester in ML and Climate at Columbia University during the Fall of 2022.

I develop an approach to help identify promising locations to build solar panel arrays by leveraging only GHI levels and ambient temperature readings from the National Renewable Energy Laboratory (NREL)'s satelitte data to approximate the historical potential Photovoltaic output that matches SOTA simulations. I explore machine learning models from Linear Regression to Random Forest approachs and document the results in notebooks viewable in this repository.

First, run requirements.txt:

$ pip install -r requirements.txt

Note: You will also need to install osgeo (GDAL) to process the raster files. A .whl file is included to install the version used in this project

To walkthrough this project, view the following notebooks under the src/Notebooks folder in sequence:

  • 1 Map Visualization and Data Selection
  • 2 EDA on DATA
  • 3 Data and Modeling

My process is documented in journal.md and my resources are listed in resources.md

About

ML Climate Project

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 98.4%
  • Python 1.6%