A new approach to determine the long-term effect of efficiency losses due to different dust types accumulation on PV modules with artificial neural networks

Photovoltaic modules are faced with many kinds of environmental effects. One of the most important factors on the efficiency of these systems is dust accumulation on the module surfaces. Airborne particles of dust from various sources, gradually accumulates on the surface of PV modules, thereby lowering the incident radiation reaching to the solar cells. The formation of dust accumulation changes depending on the air temperature, humidity, wind patterns, soil structure, and vegetation characteristics of the region where these systems are installed. Although there are many studies focused on the short-term and medium-term effects of dusting on photovoltaic modules, the long-term studies are quite limited due to some involved problems. In the present study, to fill this gap, the effects of long-term dust accumulation amounts were determined by interpreting the effects of short-term dust accumulation from different origins with a developed artificial neural network model. In the study, clay, calcareous and urban dust samples were used to reveal the effect of different dust sources. Although the effect of dusting on efficiency was given over time, the amount of dust accumulated on the module surface was also expressed in mass in order to find out a general insight. Based on the findings, after 1 month of exposure to clay, calcareous, and urban dusts, the solar modules experienced efficiency losses of approximately 18%, 26%, and 28% respectively, when compared to clean modules. Subsequently, at the end of 3 months, the efficiency losses increased to approximately 46%, 61%, and 63% respectively, and after 6 months, the losses reached approximately 67%, 77%, and 78% respectively. All dust types, particularly urban and calcareous dust, led to substantial efficiency losses in photovoltaic modules. Consequently, regular cleaning of the modules is recommended to prevent these losses.