Deep learning‐based substation remote construction management and AI automatic violation detection system

Abstract

Workplace video surveillance and timely response to operational violations are critical to avoid operator injuries at power construction sites. Here, a system that combines remote substation construction management and artificial intelligence object detection techniques to intellectualize the power construction management process and identify violations during construction in real time is proposed. To improve the detection accuracy, a data augmentation method, including three operations: (1) object segmentation and background fusion; (2) partial erasing; and (3) other basic transformations, is also proposed. Six variants of the You Only Look Once (YOLO) model are trained as detectors for comparative experiments on a dataset collected at the practical construction site. The experimental results show that the detection precision and recall of the YOLOv5‐s model are 0.852 and 0.922, with high accuracy and low miss rate, which meet the requirements of robustness and accuracy in detecting realistic power construction violations.

An Identification Method for Irregular Components Related to Terminal Blocks in Equipment Cabinet of Power Substation

“…To cope with the challenge from the data source, a series of data enhancement operations are applied. Along with the subdivision of the training, validation, and testing datasets, numerous basic online enhancement operations (color distortion, image flipping and rotation, and noise addition) [ 10 ] are used during the training stage. Moreover, a mosaic data enhancement strategy involving image slicing, cropping, and combination [ 27 ] is utilized for the block area location model, which allows the area detection model to function well in datasets of a smaller scale.…”

“…Similarly, Zhao et al [ 9 ] proposed a personal safety protective equipment detection model to remind employees who violated safety regulations during operations to take action, protecting them from electrical injuries. Yan et al [ 10 ] compared six variants of the You Only Look Once (YOLO) [ 11 , 12 , 13 ] model, which served as object and violation detectors in the realistic substation construction site of different scenarios. Thanks to the algorithms’ open-source nature, the researchers face fewer obstacles to reproducing and applying the detection model and can focus on the adjustments and improvements to the concrete problems.…”

Optimized hybrid YOLOu‐Quasi‐ProtoPNet for insulators classification

“…YOLO is a CNN that splits images into grids, having each grid cell detect objects within itself [51]. This approach is a single‐shot algorithm, which means that it only requires processing the image once to detect and classify the object under consideration.…”

Particle swarm optimization method for substation location of airport surrounding power grid considering civil aviation clearance limit surface

“…In recent years, with the development of aviation industry and the increase of airport operation, the civil aviation clearance around the airport has become an important issue [1]. As a guarantee of flight safety, the clearance limit surface around the airport ensures the safe distance of the aircraft in the process of taking off and landing, and at the same time ensures that nearby buildings and facilities will not interfere with the process of taking off and landing [2][3]. However, when determining the location of power grid substations around the airport, it is necessary to consider the relationship between the clearance limit surface and power grid facilities to ensure flight safety and efficient and reliable power supply.…”