Research on detection method of photovoltaic cell surface dirt based on image processing technology

Area of Science:

• Renewable Energy
• Computer Vision
• Photovoltaic Technology

Background:

• Reduced power generation efficiency due to photovoltaic panel soiling.
• Limitations of manual inspection methods: high workload and low efficiency.

Purpose of the Study:

• To develop an automated method for detecting dust and dirt on photovoltaic panels using image processing.
• To enhance power generation efficiency by enabling timely detection and elimination of soiling-related issues.

Main Methods:

• Image processing techniques including mathematical morphology (enhancement, sharpening, filtering, closing operations).
• Image histogram equalization to improve contrast and detail under uneven illumination.
• Dirt area proportion calculation for fault identification.
• Improved A* path planning algorithm for efficient unmanned aerial vehicle (UAV) inspection.

Main Results:

• Enhanced image quality for clearer dirt identification.
• Effective binarization despite uneven illumination.
• Accurate fault detection based on dirt area ratio.
• Significantly improved efficiency of UAV-based photovoltaic panel inspection.

Conclusions:

• The proposed image processing and A* algorithm method effectively detects photovoltaic panel soiling.
• This automated approach enhances inspection efficiency, reduces operational costs, and improves overall photovoltaic unit maintenance.