Detection of Suboptimal Conditions in Photovoltaic Systems Integrating Data from Several Domains

Leonardo Cardinale-Villalobos, Luis D. Murillo-Soto, Efrén Jimenez-Delgado, Jose Andrey Sequeira

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

Abstract

Researchers have been exploring methods to detect suboptimal conditions in photovoltaic (PV) modules, such as visual inspection, electrical analysis, and thermography. Each method has its advantages and limitations. To enhance the accuracy and efficiency of detecting these conditions, this research proposed an integrated Internet of Things (IoT) platform called the Multi-method system (MMS). This platform combines Infrared Thermography (IRT), visual inspection using RGB image processing, and electrical analysis. The MMS enables automated data capture, analysis, and remote accessibility. To validate the system, an experiment was conducted at the Costa Rica Institute of Technology. The MMS effectively detected suboptimal conditions due to soiling, partial shading, and short circuits. The system achieved a sensitivity of 0.97%, an accuracy of 0.98%, and a precision of 100%. This project functions as a proof of concept, in this case limited to a fixed location and for specific suboptimal conditions, however, it represents a solution with potential for industrial use. The research contributes to advancing PV system reliability and performance monitoring in Smart Cities, offering implications for improving solar energy efficiency and reducing maintenance costs.

Original languageEnglish
Title of host publicationSmart Cities - 6th Ibero-American Congress, ICSC-Cities 2023, Revised Selected Papers
EditorsSergio Nesmachnow, Luis Hernández Callejo
PublisherSpringer Science and Business Media Deutschland GmbH
Pages18-32
Number of pages15
ISBN (Print)9783031525162
DOIs
StatePublished - 2024
Event6th Ibero-American Congress on Smart Cities, ICSC-Cities 2023 - Cuernavaca, Mexico
Duration: 13 Nov 202317 Nov 2023

Publication series

NameCommunications in Computer and Information Science
Volume1938 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference6th Ibero-American Congress on Smart Cities, ICSC-Cities 2023
Country/TerritoryMexico
CityCuernavaca
Period13/11/2317/11/23

Keywords

  • CNN
  • Deep Learning
  • Fault Detection Technique
  • Internet of things platform
  • Photovoltaic

Fingerprint

Dive into the research topics of 'Detection of Suboptimal Conditions in Photovoltaic Systems Integrating Data from Several Domains'. Together they form a unique fingerprint.

Cite this