Only 39% of electricity sales come from residential areas, while 71% of electricity sales come from commercial and industrial sectors. To make the greatest impact possible, I wanted to find a way to increase the use of renewable energy in the energy industry. Although the United States government spends over 15 billion dollars on sustainable energy every year, most sustainability projects end in financial failure. This is because of 3 main reasons: they choose the wrong location, can't get investors, and the government fails to support them. This all stems from one major issue: they do not undersand the demand for renewable electricity.

What it does

Clevergy is a web app which promotes the understanding of electricity demand over time. It helps inform governent budgeting and large insitutional investors to increase the understanding of electricity consumption over time. This helps them to target investments where they are really needed, optimizing the use of financial resources in the renewable elecrticity industry.

How we built it

  1. Clevergy accesses raw historic data from the EIA, the US Energy Information Administration about net electricity generation by sector.
  2. Python uses Facebook Prophet inorder to detect yearly trends and monthly patterns.
  3. Facebook Prophet trains on EIA Time Series data from 2016 to 2020 and predicts values from 2021 to 2025. The predictions for 2021 to 2022 are then validated against the actual data reported by EIA with high accuracy averaging around 92%.
  4. Python plots and saves a visual representation of this on the Clevergy website for human consumption.

Challenges we ran into

I tried to use Facebook Prophet on price data as well, however becuase of recent events such as the Russia-Ukraine war and COVID-19, the data was too eratic and my accuracy was very low. I had to give up on this idea, however I plan to keep learning about price analysis to make this possoble one day.

I hope that Clevergy can be used in order to disuade the government from passing bills such as the Willow Project act currently in congress. We should be using past data to predict future impacts of such decisions.

Share this project: