The accumulation of dust on the photovoltaic (PV) modules for the PV plant result in energy loss and therefore, there was a need to assess the impact of soiling on the PV modules since soiling losses are location specific. In this work, the approach of quantifying soiling loss experimentally which involved comparing the temperature corrected short circuit current values for the naturally soiled string to the frequently cleaned string had higher accuracy than the method of extracting soiling loss from the PV plant production data. This was due to the uncertainties associated with the plant production data.
The average soiling loss values were determined experimentally and from the plant production data. The month of January 2018 was found to have the highest soling loss of 6 % and March 2018 had the lowest soiling loss was less than 1 %. The soiling for the month of January 2018 was used to determine an optimum cleaning interval which balances out with the revenue lost due to soiling and the cost of the cleaning event using the cleaning schedule model and the optimum interval was 21 days from the last date when the plant last cleaned. This optimum cleaning interval reduces the total cost per unit energy generated by the PV plant and therefore, the levelized cost of electricity. However, the optimum cleaning interval may vary depending on the cost of the cleaning event as well as the seasonal variations in the soiling loss and energy generated. The cleaning schedule model developed can be used to determine when the PV plant should be cleaned. However, this model should be used as a guideline since soiling loss highly depends on the climate of the area and it is always changing.