Design of an energy-aware and optimization-based intelligent PV cleaning system

In this thesis, two fundamental challenges that reduce energy production efficiency in photovoltaic (PV) systems are addressed through an integrated approach: (i) modeling and prediction of power losses caused by surface soiling, and (ii) optimization of energy and resource consumption in panel cleaning processes. Most studies in the literature assume that contamination on PV panels is uniformly distributed; however, such an assumption does not adequately represent real-world performance degradation under field conditions. In the first stage of this study, a novel Transmittance Distribution (TD) approach was developed to physically model the effect of non-uniform contamination on the output characteristics of PV modules. The TD extraction was carried out using the Equivalent Circuit Model (ECM) combined with optimization-based solution techniques, and the results were validated against experimentally measured I–V curves with high accuracy. The proposed method is capable of explaining nonlinear behaviors influenced by environmental factors such as irradiance and temperature and, unlike conventional models, can accurately predict the actual power loss of panels under non-uniform soiling conditions. In the second stage, cleaning decision-making and path planning were investigated. For this purpose, a new energy-aware route optimization method based on the Recursive Tabular Validation Algorithm (RTVA) was developed. The RTVA algorithm optimizes the cleaning sequence and robotic motion path by considering partially soiled coordinates detected through image processing (IP), thereby minimizing unnecessary resource consumption. Together, these two approaches—TD optimization and RTVA path planning—form an integrated decision-support system that spans the entire process from deriving the contamination distribution based on measured efficiency loss to predicting future energy loss and determining the optimal cleaning strategy. The findings contribute to improving energy production efficiency, enhancing cost-effectiveness of cleaning operations, and promoting sustainable operation strategies for PV systems.