Remote Visualization and Automatic Inspection Method of Smart Substation Based on Surveillance Video Image
Due to the interference of various magnetic fields in the data collection of substations,the visual inspection results obtained by traditional inspection methods are relatively vague.To solve this problem,this paper proposes a remote visual automatic inspection method for smart substations based on monitoring video images.Firstly,establish a video image dataset for substation monitoring,expand the dataset using image data augmentation technology,identify key equipment and areas in the substation through inspection positioning technology,and finally,implement inspection image recognition based on the YOLOv3 network framework to achieve visual automatic inspection function.The experimental results show that the accuracy of the inspection results using the proposed method is higher than 95.3%,which is better than the comparison method and has a good inspection effect.
surveillance video imagesmart substationremote visualizationvisualization and automatic inspection
NETL
NSTL
万方数据
