/Predicting-Power-Flow

Final project for EECE E6895 - Predicting Power Flow using ML Techniques

Primary LanguageJupyter Notebook

Final Project for Advanced Big Data Analytics - EECS E6895

Power Flow Optimization using Big Data Techniques

Rohan Raghuraman (rr3417@columbia.edu)

Project Description

The goal of this project is to optimize the power flow in a region under various conditions and to determine if the optimal power flow problem can be made faster using machine learning and AI. This is an important task because increased energy demand, addition of renewable energy and electric vehicles had led to increased challenges in terms of maintaining system stability and economic viability. Suboptimal power flow conditions could exacerbate climate change. The end goal is that the techniques in this project will contribute to lowering the time and computing energy needed to perform power flow optimization so that the development of the energy landscape will not be hindered.

To setup, first, clone the repository.

To run the web dashboard and generate predictions, follow the below steps:

  1. Open the Web Dashboard file.
  2. Open app.py.
  3. Under @app.route('/chart1'), change the directory of the df to wherever your all_forecast.csv is stored.
  4. Run the file.
  5. Open the Flask url in your localhost.
  6. Play around!