A Novel Single-Axis Solar Tracker Based on Reinforcement Learning

Ming Huang*, Ziqiang Bi, Jieming Ma, Xiaohui Zhu, Jie Zhang, Ka Lok Man

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

1 Citation (Scopus)

Abstract

Solar panels with fixed angles can hardly consistently achieve maximum power output because the sun's rays are not always perpendicular to the panel. To address this problem, a single-axis solar tracker based on reinforcement learning (RL) is designed in this paper. The proposed RLbased solar tracker has a simple structure that no additional sensors or positioning devices are required. Therefore, the maintenance of the proposed tracker is easy to perform. Simulations have been conducted to validate the effectiveness of the proposed solar tracker. The results demonstrate that the mean absolute error (MAE) between the tracking tilt angle and the theoretical value is within a one-eighth degree.

Original languageEnglish
Title of host publicationProceedings - 2022 International Conference on Industrial IoT, Big Data and Supply Chain, IIoTBDSC 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages329-333
Number of pages5
ISBN (Electronic)9781665454551
DOIs
Publication statusPublished - 2022
Event3rd International Conference on Industrial IoT, Big Data and Supply Chain, IIoTBDSC 2022 - Virtual, Online, China
Duration: 23 Sept 202225 Sept 2022

Publication series

NameProceedings - 2022 International Conference on Industrial IoT, Big Data and Supply Chain, IIoTBDSC 2022

Conference

Conference3rd International Conference on Industrial IoT, Big Data and Supply Chain, IIoTBDSC 2022
Country/TerritoryChina
CityVirtual, Online
Period23/09/2225/09/22

Keywords

  • reinforcement learning (RL)
  • simulation
  • single-axis tracker
  • solar panel
  • solar tracker

Fingerprint

Dive into the research topics of 'A Novel Single-Axis Solar Tracker Based on Reinforcement Learning'. Together they form a unique fingerprint.

Cite this