A novel fault detection technique for PV systems based on the K-means algorithm, coded wireless Orthogonal Frequency Division Multiplexing and thermal image processing techniques

Solar energy is considered one of the most ecological energy systems, which provides clean, reliable, and un-limited power. However, the solar system can be exposed during its operation to numerous failures that can immediately lead to a dangerous incident such as fire ignition. To ensure the productivity and safety of the system, it must be accompanied by fault detection and diagnostic (FDD) techniques. A new wireless commu-nication system and its application in PV systems to detect damaged areas of solar panels are proposed to address these issues. The proposed approach is composed mainly of two parts; the first part aims to check and detect the existence of damaged areas in the thermal images of PV panels, while the second part focuses on the transmission of faulty panels using Orthogonal Frequency Division Multiplexing (OFDM) modulation with three decoding algorithms, namely, Min-sum, Bit-Flipping, and Viterbi, to send and correct errors induced by a wireless transmission channel. The simulation results have shown that employing the K-means algorithm, OFDM mod-ulation, and the Min-sum decoding scheme offers a significant performance improvement giving a rapid degradation with a BER of 10(-4) for SNR = 3.7 dB and a null BER from SNR = 4 dB. From a financial point of view, it was proved that the adopted solution is cost-effective compared to literature-based ones. This solution could be very useful since the overall capital cost is about 5.41 USD/m(2).

Automated summary

This article introduces a new wireless communication system and its application in PV systems to detect damaged areas of solar panels. By using specific algorithms and modulation schemes, rapid detection and error correction of faulty panels are achieved. The research results demonstrate significant advantages of this solution in terms of performance and cost-effectiveness.