Yangquan Photovoltaic Panel

Can deepsolar be used to build a PV installation database?

Yu et al. have made important progress by proposing DeepSolar, a deep learning framework that applied to 0.3 m satellite imagery for constructing a PV installation database, their evaluation has demonstrated that DeepSolar can achieve significantly high accuracy on PV panel segmentation and size estimation.

How to monitor the deformation of substations and photovoltaic power stations?

In this paper, the time series SAR images of the 25 Sentinel-1A satellite in Yangquan from January to November 2019 are obtained to monitor the deformation of substations and photovoltaic power stations by using the SBAS-InSAR technology. The results show that the deformation range of the study area is between -23.8cm and 11.2cm.

How accurate is the assessment of PV potential & capacity in complex scenes?

The accurate assessment of PV potential and capacity in complex scenes requires the detailed surface information to be observed and parametrized more accurately.

Can a deep siamese network identify solar panels from low-resolution historical images?

Wang et al. have used a deep siamese network to identify solar panels from low-resolution historical images by learning the correlation between the historical images and the high-resolution exemplar images.

How accurate is PV fault detection from RGB images based on deep learning?

Despite the advancement achieved by these studies, PV fault detection from RGB images based on deep learning is still facing challenges in detection accuracy and adaptability to different defect categories, largely because of the limited amount of labeled dataset and insufficient transferability of models.