Predicting solar photovoltaic electrical output under variable environmental conditions: Modified semi-empirical correlations for dust

The current study investigated the simultaneous impacts of dust as well as other environmental parameters on cell temperature and the electrical outputs of photovoltaic systems. Based on the experimental and analytical analysis, several linear and polynomial semi-empirical correlation forms were proposed. It was shown that cell temperature depends directly on irradiation and ambient temperature, as well as inversely on humidity, wind speed, and the amount of accumulated dust on the surface. During the four-year outdoor experiments and fur-ther indoor measurements, the surface dust density varied from 0 to 10.1174 g.m-2, which resulted in up to a 28 % current loss. Although, it was shown that dust affects the output voltage less, and the most recorded varia-tion was <0.5 % reduction. The electrical behavior of the photovoltaic system was also simulated with a modified single-diode model. Comparing the simulation results, correlation predictions, and experimental measurements confirmed consistency. Moreover, it was shown that considering the prediction precision, simplicity, and compu-tational cost, the proposed correlation forms should be preferred when just the output current, voltage, and power are of concern. The proposed modified semi-empirical correlation forms can be easily coupled with other models to improve their precision and accuracy. (c) 2022 International Energy Initiative. Published by Elsevier Inc. All rights reserved.

Automated summary

The current study investigates the simultaneous impacts of dust and other environmental parameters on cell temperature and electrical outputs of photovoltaic systems. Linear and polynomial semi-empirical correlation forms are proposed based on experimental and analytical analysis. It is shown that cell temperature is directly influenced by irradiation and ambient temperature, as well as inversely influenced by humidity, wind speed, and accumulated dust. The study demonstrates the effect of dust on current loss, with a variation of up to 28%, while the impact on output voltage is minimal (<0.5% reduction). The proposed correlation forms provide accurate predictions for output current, voltage, and power, with simplicity and computational efficiency. They can be easily integrated with other models to improve precision and accuracy.