A temporal-spatial cleaning optimization method for photovoltaic power plants

Cleaning is critical for photovoltaic (PV) systems, as it can remove dust deposition and keep the systems operating efficiently. Existing studies on PV cleaning focus predominately on determining optimal cleaning frequencies or intervals. In practice, however, route planning is also an indispensable decision to be made when cleaning large-scale PV plants. In this work, we study a temporal-spatial cleaning optimization problem for large-scale PV plants. A two-stage cleaning optimization policy that consists of a periodic planning stage and a dynamic adjustment stage is proposed. In the former stage, tentative cleaning intervals are determined; in the latter stage, the temporal scheduling and spatial routing problems are jointly optimized in a dynamic fashion, so as to minimize the total economic loss. We model the problem as an extended version of the classical traveling salesman problem (TSP), that is, a production-driven traveling salesman problem with time-dependent cost (PD-TSP-TC). Genetic algorithm is employed to solve the corresponding nonlinear 0–1 integer programming problem. A case study on a real PV plant is conducted to demonstrate the performance of the proposed method, which shows that the temporal-spatial cleaning optimization method results in a better performance than the one that ignores route planning and that models route planning as classical TSP.