Designing and Implementing a Novel Solar Panel Tracker System Leveraging Reinforcement Learning Technique

Overview

Objectives As of today, the average efficiency of household solar panels is around 20%, which means only around 20% of solar energy is converted into household electricity. Our objective is to explore various techniques that not only enhance this efficiency but also are scalable for thousands of households. Methods Designed and constructed an intelligent solar panel tracker that adjusts the solar panel s orientations to maximize the sun ray receptance throughout a day, intelligence offered by a reinforcement learning technique that explores an optimal orientation, taking into account of the variations in the environmental factors. Placed two identical solar panels side-by-side, one mounted on our solar panel tracker; the other has a fixed position. Collected voltage measurements from both solar panels at identical time intervals between 8 AM to 5 PM each day. Repeated the same benchmark testing for several days. Tracker system mainly consists of servo motors, Arduino board, rotational platform, and solar panel. Results Five trials were conducted. Daily V oltage Output Percentage Gains: 1st trial-11% ; 2nd trial-16% ; 3rd trial-22% ; 4th trial-16% ; 5th trial-21%. Average percentage gain is 17%. The positive voltage output gains show that this approach is effective. Conclusions This is the first ever solar tracking system that leverages a reinforcement learning technique. Our experiment data consistently shows positive voltage output percentage gains from the smart solar tracker throughout our trials. With on average nearly 20% of efficiency gain, it clearly shows that our novel approach is effective. Based on the data charts, the efficiency percentage gains in the morning and evening are greater than the daily peak temperature time window, usually centered around noontime. Also, on a warmer day, the daily efficiency percentage gain tends to be lower. This is because the fixed solar panel is positioned facing up vertically, thus likely to output close to optimal voltages during peak temperature hours. In addition, a warmer day also tends to have longer peak temperature time window. As the temperature or light intensity goes higher, both smart and baseline solar panel voltage outputs are higher, so should expect the percentage gain from tracker be lower.

Awards (1)

• Honorable Mention

Competition history

• CSEF 2019 Alternative Energy (Junior Division Only) · Entry J0210

Resources

• CSEF project summary ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Energy and Transport projects
• CSEF Energy and Transport projects
• Every CSEF 2019 project