| General note |
The health of solar PV systems is vital for maintaining efficiency, reliability, and<br/>safety. Although PV systems are inherently low-maintenance and lack moving parts, their<br/>performance depreciates over time due to environmental stressors such as dust<br/>accumulation, temperature fluctuations, microcracks, and panel misalignment. Effective<br/>monitoring systems enable the early detection of these issues, helping timely initiating the<br/>maintenance procedures, improving energy production, and extending system lifespan.<br/>This is of particular significance since improving PV efficiency through proactive<br/>management is often more cost-effective than expanding capacity through additional<br/>investments. This study presents an innovative approach for evaluating solar panel health<br/>using ML-driven analysis, as an alternate of traditional multi-sensor methodologies. The<br/>proposed technique extracts multiple attributes from a single data stream, employing ML<br/>algorithms to predict environmental impacts on PV performance. Sensor data was analysed<br/>to find key characteristics that accurately represent various aspects of panel health.<br/>Initially, temperature effects on a 100 kW grid-tied solar system deployed across<br/>various installation sites, including parking lots, rooftops, and ground-level arrays were<br/>examined. The results indicated that variations in operational temperature led to power<br/>losses of 27.95% in ground-mounted systems, 5.41% in rooftop systems, and 0.82% in<br/>parking-based systems. These outcomes were derived from temperature measurements at<br/>the three installation sites, further supported by weather station data. TMR sensors were<br/>also investigated for their ability to check multiple PV panel variables, particularly threeaxis orientation. Experimental findings revealed that the largest axis error values for Bx,<br/>By, and Bz, the magnetic field components along the X, Y, and Z axes, were below 0.1%,<br/>0.3%, and 0.5%, respectively. However, TMR sensors proved insufficient for accurately<br/>capturing solar panel temperature fluctuations, prompting a shift toward EIS for a more<br/>comprehensive analysis. EIS experiments were conducted under controlled temperature<br/>conditions ranging from 10°C to 60°C, while maintaining all other environmental variables<br/>constant. The EIS response was recorded for 26 distinct parameters, providing a detailed<br/>dataset for analysis. The 26 distinct parameters are various electrical values obtained from<br/>the EIS experiments. ML classification of the EIS data using Caret-based techniques foundxvii<br/>the Light Gradient Boosting Machine and CatBoost Classifier as the more suitable models,<br/>achieving classification accuracies of 85.63% and 85.54%, respectively. Additionally, EIS<br/>was used to detect microcracks and dust accumulation under four real-world scenarios: (1)<br/>clean and healthy panels, (2) dusty but healthy panels, (3) clean panels with microcracks,<br/>and (4) dusty panels with microcracks. The Gradient Boost classifier achieved an<br/>impressive 97.5% accuracy in distinguishing these conditions. The experimental results<br/>indicated that combining EIS with machine learning provides a powerful approach for<br/>assessing solar panel temperature variations, detecting microcracks, and monitoring dust<br/>accumulation through DC power lines alone. This method demonstrates considerable<br/>potential for real-time PV system diagnostics, thereby enhancing predictive maintenance<br/>and operational monitoring. |
