Collaborative Control Framework of IoT-Enabled Intelligent Cleaning Systems in Photovoltaic Stations

This paper addresses the challenges of collaborative control in an IoT-based intelligent cleaning system for photovoltaic power plants. The paper designed an overall architecture encompassing the perception, network, and application layers. The paper proposed a multi-agent collaborative mechanism and a collaborative control strategy based on rules and optimization algorithms. We investigate key technologies, including photovoltaic panel contamination detection, intelligent control of cleaning equipment, and data fusion processing. Using a MATLAB/Simulink simulation environment, the paper compared the performance of traditional control with this framework under varying pollution levels, weather conditions, and equipment loads. Results show that under moderate pollution conditions, the framework reduces task completion time by 32.4%, improves resource utilization by 22.3 percentage points, and reduces equipment conflict by 67.5%. Furthermore, under light rain conditions, the power generation efficiency improves by 8.7 percentage points compared to traditional control. These results demonstrate its advantages in cleaning efficiency, resource utilization, and environmental adaptability, supporting the efficient operation of photovoltaic power plants.