Rapid access to the operating status of Photovoltaic (PV) panels and troubleshooting can save management and maintenance costs for the development of PV power plants, which is important for PV power plant management and power generation capacity assurance. The use of remote sensing technology to identify the faults of photovoltaic panels has developed rapidly, however, current research usually relies only on a single optical data source to identify and count the area of PV panels in a PV electric field, although there are literature on PV panel fault detection, only the surface fault identification of PV panels is tested, while the internal faults (such as panel bad points or bad lines) cannot be identified because of the limitations of optical remote sensing. In this paper, a photovoltaic panel fault monitoring technology based on multi-source remote sensing is proposed. The optical and thermal infrared hybrid data combined with deep learning technology are used to achieve rapid and accurate fault identification and localization of PV panel arrays. It can not only automatically identify PV panels that are obscured by dust and foreign objects, but also locate PV panels that have bad dots or bad lines, which greatly improves the ability and effectiveness of remote sensing PV panel fault monitoring. The high-resolution unmanned air vehicle (UAV) optical image and thermal infrared image are applied in this experiment. The Mask RCNN algorithm is used to accurately locate and number the photovoltaic panel of the optical image. Then, the fault scene classification model is established for the multi-type fault characteristics of the optical image and thermal infrared image within the panel range, so as to identify five types of faults, such as dust cover, branch cover, bird droppings cover, internal bad points and bad lines of PV panel, which effectively solves the problem that the single optical remote sensing image cannot identify the internal component faults of the photovoltaic panel under normal conditions.

The spatial distribution of Ground Control Points (GCPs) is a critical factor affecting the accuracy of UAV photogrammetry in hilly terrain. Existing studies primarily focus on the influence of GCP quantity on accuracy or discuss planar distribution uniformity in flat areas, with limited systematic analysis separating horizontal and vertical distributions as independent dimensions. This study utilizes a DJI Mavic 3E UAV to acquire aerial imagery in a typical hilly area of Mengyin County, Shandong Province, with 27 high-precision GCPs deployed. Four comparative experiments combining random/uniform distributions in both horizontal and vertical dimensions are designed to quantitatively analyze the impact of different distribution patterns on aerial triangulation and mapping accuracy. Results demonstrate that the dual-uniform distribution strategy achieves optimal accuracy, with horizontal RMSE of 0.045 m and vertical RMSE of 0.039 m, representing improvements of 32.8% and 17.0% respectively compared to the random distribution scheme. Furthermore, this paper proposes the Spatial Distribution Balance Index (SDBI), which integrates Planar Uniformity Index (PUI) and Vertical Uniformity Index (VUI) with a terrain-adaptive weighting mechanism. The VUI weight, calibrated as β=0.714 via a Sigmoid nonlinear amplification function (k=15,x0=0.15), enables the SDBI to adaptively reflect terrain sensitivity to vertical control. The enhanced SDBI shows a correlation coefficient of r=−0.93 with final accuracy, validating its effectiveness as a GCP layout optimization and evaluation tool. This research provides a quantifiable technical framework for GCP deployment in UAV photogrammetry in hilly regions.

The spatial distribution of ground control points (GCPs) is a critical factor affecting the accuracy of UAV photogrammetry in hilly terrain. In existing studies on GCP distribution, researchers have largely focused on planar uniformity metrics in flat terrain or on the effects of flight parameters in mountainous areas, with limited attention to the distinct roles of horizontal and vertical placement. In this study, we utilized a consumer-grade RTK-equipped UAV to acquire aerial imagery in a typical hilly area, with 27 high-precision GCPs deployed as a reference dataset. Four comparative experiments combining random/uniform distributions in both horizontal and vertical dimensions were designed to quantitatively analyze the impact of different distribution patterns on aerial triangulation and mapping accuracy. Our results demonstrate that the dual-uniform distribution strategy (i.e., uniform in both planimetric layout and elevation stratification) achieves the highest accuracy among the four tested configurations, with horizontal RMSE of 0.045 m and vertical RMSE of 0.039 m. Furthermore, we propose the Spatial Distribution Balance Index (SDBI), which integrates the Planar Uniformity Index (PUI) and Vertical Uniformity Index (VUI) with a terrain-adaptive weighting mechanism. The VUI weight, exemplified as β = 0.714 for this study area via a Sigmoid nonlinear amplification function (k = 15, x₀ = 0.15), enables the SDBI to adaptively reflect terrain sensitivity to vertical control. The enhanced SDBI exhibits a correlation coefficient of r = −0.93 with final accuracy, validating its effectiveness as a GCP layout evaluation tool. In this study, we establish the SDBI as a diagnostic metric that quantitatively links GCP distribution characteristics to photogrammetric accuracy outcomes, providing both theoretical insights into anisotropic error propagation and practical guidance for deployment design in hilly regions.