Solar Photovoltaic Array Fault Diagnosis Using an IoT and Semi-Supervised Learning-Based Strategy

Electrical Engineering

Faults inside the arrays of solar photovoltaic (SPV) systems have a major impact on their performance. The nonlinear properties of solar cells and changes in the environment make it difficult for traditional protection techniques to detect flaws. Large amounts of data and sensors are needed for learning-based fault diagnosis techniques, which makes them expensive and difficult to integrate with current SPV systems. Furthermore, current methods are unable to locate the problematic panel precisely due to the absence of panel-level monitoring. To tackle these problems, this paper suggests a methodology for fault detection, pinpointing the precise damaged panel, and precisely categorizing errors in SPV arrays. A sensor less electronic circuit with a bipolar junction transistor (BJT) and Zener diode for fault detection, an Internet of Things (IoT)-based web application for tracking individual solar panels for fault localization, and a semi-supervised learning module based on a deep auto encoder (DAE) followed by a hybrid support vector machine and logistic regression (SVM-LR) for fault classification are all part of the methodology.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.