Multidefect Detection Tool for Large-Scale PV Plants: Segmentation and Classification

Unmanned aerial vehicles (UAVs) with high-resolution optical and infrared (IR) imaging have been introduced in recent years to perform inexpensive and fast inspections in operation and maintenance activities of solar power plants, reducing the labor needed, while lowering the on-site inspection time. Even though UAVs can acquire images extremely quickly, the analysis of those images is still a time-consuming procedure that should be performed by a trained professional. Therefore, a computer vision approach may be used to accelerate image analysis. In this work, a dataset of IR images was created from a 10-MW solar power plant and a comparative analysis between mask R- convolutional neural network (CNN) and U-Net was performed for two experiments. Concerning the defective module segmentation, the mask R-CNN algorithm achieved a mean average precision at intersection over union (IoU) = 0.50 of 0.96, using augmentation data. Regarding the segmentation and classification of failure type, the algorithm reached a value of 0.88 considering the same evaluation metric and data augmentation. When compared to the U-Net in terms of IoU, the mask R-CNN outperformed it with 0.87 and 0.83 for the first and second experiments, respectively.

Automated summary

Unmanned aerial vehicles (UAVs) with high-resolution optical and infrared (IR) imaging have been used for fast and cost-effective inspections in solar power plants. However, the analysis of the acquired images is still time-consuming and requires trained professionals. This study compares the performance of mask R-CNN and U-Net for image analysis in a 10-MW solar power plant, achieving high precision and outperforming U-Net in terms of IoU.