Building Integrated Solar Photovoltaics Assessment via ML

Partner: Sergio Castellanos, UC Berkeley, Academic

Overview

Project Description

With the advent of novel image classification algorithms, the opportunities to apply them to the energy space are increasing, too. In this project, we are looking to expand these capabilities into an ongoing task of assessing the potential of building integrated solar photovoltaics through ML image classification. You’d be given a (limited) dataset, with the opportunity to expand it beyond and implement different algorithms. Beyond that, you’d also venture into the energy space by modeling the operation of PV systems and link both the ML field with the energy space.

Expected Deliverable

Model estimates, both in classification performance as well as in final energy production.

What would a successful semester look like to you?

Student has challenged itself by taking ownership of the project and printing its creativity.

Data

Models

Conclusion