Skip to content

2nd Place Project from 2023 MIT Energy and Climate Hack

Notifications You must be signed in to change notification settings

owenfrausto/carbon-cut

Repository files navigation

Carbon Cut

An open-source resource for making high-quality building decarbonization characteristic predictions from simple and common data points, found in the NREL ResStock dataset

This repository serves as documentation for our project, and a guide for future work. This project was started as part of the WattCarbon Challenge at the MIT Energy & Climate Hackathon in November 2023, where it won the second place prize. Presentation here, starting at 56:00.

-> savings_prediction.ipynb trains and evaluates a model to estimate the savings (in terms of carbon emissions, energy, and dollars) for each intervention in a given set, given only the hourly loadshape data for the building in question.

-> metadata_prediction.ipynb aims to train a model to predict specific and relevant aspects of a building (eg. square footage, foundation type, appliance efficiency) given the hourly loadshape.

About

2nd Place Project from 2023 MIT Energy and Climate Hack

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published