Quantitative Analysis of Solar Photovoltaic Panel Performance with Size-Varied Dust Pollutants Deposition Using...

Area of Science:

• Renewable Energy
• Materials Science
• Artificial Intelligence

Background:

• Solar photovoltaic (PV) panels are crucial for renewable energy generation.
• Dust accumulation on PV panels is a major environmental challenge affecting performance.
• Understanding the impact of dust particle size is vital for PV efficiency maintenance.

Purpose of the Study:

• To investigate the effect of different dust particle sizes on solar PV panel performance.
• To develop and compare machine learning models for predicting PV output power under dust deposition.
• To assess the suitability of machine learning for performance prediction in dusty environments.

Main Methods:

• Experimental analysis of PV panel performance with five dust sizes at 33.48 g/m² deposition.
• Utilizing machine learning regression models: Support Vector Machine Regression (SVMR), Multiple Linear Regression (MLR), and Gaussian Regression (GR).
• Comparing ML model accuracy using Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared (R²) metrics.

Main Results:

• Zero-resistance current decreased by up to 49.01% for small dust particles and 15.68% for large particles (600-850 µm).
• Sunlight penetration reduced by nearly 40% due to smaller dust particles compared to larger ones.
• SVMR demonstrated optimal performance (MAE=0.1589, MSE=0.0328, R²=0.9919), outperforming MLR and GR.

Conclusions:

• Dust particle size significantly impacts solar PV performance, with smaller particles causing greater efficiency loss.
• Machine learning models, especially SVMR, are effective tools for predicting PV power output in dusty conditions.
• The findings support the use of ML for managing PV performance in harsh, dusty environments.