부산대학교 도서관

With the rapid capacity expansion of photovoltaics (PV) power plants, it is now increasingly important to efficiently operate these systems. One of the operational challenges is the cleaning and maintenance to mitigate yield-degradation due to soiling. While large PV plants have their cleaning strategies, these may not be optimal. It is important to understand the cleaning periods to maximize revenue and profit, and the economic losses involved away from the optimal. In this paper, we present energy and economic analysis under various periodic cleaning cycles across Bangladesh. The soiling rates are estimated from our long-term experiments in Dhaka. We also present a rain and soiling-aware, location-specific PV output model to predict energy output and revenue for 6-different locations in Bangladesh. We find that, across these locations, the profit of a solar farm can decrease by 22-35% if the panels are cleaned every 1.5 months (~50 days) instead of the optimal cycles.