College - Author 1
College of Engineering
Department - Author 1
Computer Engineering Department
Degree Name - Author 1
BS in Computer Engineering
College - Author 2
College of Engineering
Department - Author 2
Mechanical Engineering Department
Degree - Author 2
BS in Mechanical Engineering
College - Author 3
College of Engineering
Department - Author 3
Mechanical Engineering Department
Degree - Author 3
BS in Mechanical Engineering
Date
6-2022
Primary Advisor
Peter Schuster, College of Engineering, Mechanical Engineering Department
Abstract/Summary
Solar farms like the Gold Tree Solar Farm at Cal Poly San Luis Obispo have difficulty delivering a consistent level of power output. Cloudy days can trigger a significant drop in the utility of a farm’s solar panels, and an unexpected loss of power from the farm could potentially unbalance the electrical grid. Being able to predict these power output drops in advance could provide valuable time to prepare a grid and keep it stable. Furthermore, with modern data analysis methods such as machine learning, these predictions are becoming more and more accurate – given a sufficient data set. The purpose of the project presented in this report is to create a data set that can be used to train a machine learning algorithm to make these predictions.
We began our approach to creating the TSI with extensive research on rugged data acquisition systems. This research then paved the way for preliminary designs and prototypes. After some final refinements, we are confident that we have created a data acquisition system capable of lasting at least a week at the Gold Tree Solar Farm with no human intervention. This system is waterproof, weatherproof, and animal proof. It also features robust code to allow it to capture images for extended periods of time.
URL: https://digitalcommons.calpoly.edu/mesp/670
Included in
Data Storage Systems Commons, Digital Communications and Networking Commons, Mechanical Engineering Commons, Power and Energy Commons