Abstract
This chapter presents an efficient fault classification technique for monitoring the condition of photovoltaic (PV) modules. The proposed approach aims at early and efficient detection of fault to achieve reliable operation for solar PV modules. Initially, the thermal images of different module faults are captured and then preprocessed to train with the neural network classifier. Further, in the testing stage or while performing real‐time monitoring, an image processing algorithm developed using edge detection and Hough transform techniques is adapted. The chapter explains a block diagram of the proposed solar panel health‐monitoring system. The proposed panel surface area degradation analysis algorithm is developed under two phases. In the first phase, the solar panel localization is performed, and the feature extraction and analysis are developed. Further, in the second phase, the effect of PV panel surface area degradation is analyzed on the power output of the PV system.
Original language | English |
---|---|
Title of host publication | Fault Analysis and its Impact on Grid-connected Photovoltaic Systems Performance |
Editors | Ahteshamul Haque, Saad Mekhilef |
Chapter | 3 |
Pages | 77-110 |
Volume | 1 |
Edition | 1 |
ISBN (Electronic) | 9781119873778 |
DOIs | |
Publication status | Published - Nov 2022 |
Externally published | Yes |
Keywords
- Fault classification
- condition monitoring
- Photovoltaic panels
- Image edge detection
- machine learning