Realization and verification of deep learning models for fault detection and diagnosis of photovoltaic modules

Thumbnail Image

Files

master_Vo_Huynh_2021.pdf (9.54 MB)

URL

Journal Title

Journal ISSN

Volume Title

Sähkötekniikan korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2021-01-25

Department

Major/Subject

Translational Engineering

Mcode

ELEC3023

Degree programme

AEE - Master’s Programme in Automation and Electrical Engineering (TS2013)

Language

en

Pages

75+1

Series

Abstract

Recently, the usage of photovoltaic (PV) systems has grown exponentially. As a result, this places a tremendous pressure on the solar energy industry’s manufacturing sector to meet such high demand. Such issue subsequently brings many problems associated with the quality of PV systems, especially their most crucial component – the PV module. Currently, the fault and defect detection and diagnosis (FDD) process in PV modules during the manufacturing phase is challenging due to many factors, including, but not limited to the requirement of sophisticated imaging instruments and experienced personnel. Recent developments in deep learning have proven the feasibility of the latter in image classification and object detection. Thus, this approach can be extended to visual fault detection in electroluminescence (EL) and thermal (TM) imaging systems, which are the current fault detection practices in PV modules. This thesis aims to investigate and develop state-of-art deep learning approaches to detect and diagnose module visual faults and defects. Having achieved promising results, the author has shown that deep learning approaches can detect and diagnose module failures with high accuracy. From here, the author developed two deep learning classifiers capable of conducting this process efficiently and rapidly and proposed an outline for future work. This thesis can serve as a precursor for further development and conception of an automated quality monitoring and assessment system integrated directly into PV modules’ production line.

Description

Supervisor

Särkkä, Simo

Thesis advisor

Nguyen, Vinh-Khuong

Keywords

silicon solar cell, faults, electroluminescence imaging, deep learning, photovoltaic module, defects

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202101311764

Collections

[dipl] Sähkötekniikan korkeakoulu / ELEC