Overhead transmission line is an important channel for power transmission, directly related to the safe and stable operation of the power grid. Owing to the complex environment of the area through which the transmission line passes, foreign object intrusion has become the main reason for the line tripping and not reclosing successfully. The traditional defect detection method mainly relies on manual detection, and although this method is widely adopted, it still has several obvious limitations. These include low efficiency, high labour costs and susceptibility to human error. In recent years, the emergence of UAV-based inspection systems has revolutionised the way transmission lines are maintained. In this paper, an improved YOLOv8 model is proposed by replacing the YOLOv8 downsampling layer with the YOLOv9 downsampling layer ADown, and the YOLOv9 feature extraction block RepNCSPELAN4 in place of the C2F module of YOLOv8. It constitutes the YOLOv8-AR model. The method ensures that the detection accuracy of foreign object intrusion on transmission lines is improved while at the same time improving the regression rate of detection and reducing the computational volume. It performs better on the target detection task and is able to detect and localise target objects more accurately.

In the process of image dehazing, in order to improve the loss of texture details and boundary blurring caused by traditional dark channel prior dehazing algorithms after dehazing, and to avoid mutual interference caused by simultaneous processing of image structure and texture, this article proposes to layer the structure and texture of foggy images, and then optimise the algorithm separately for the defects of traditional dehazing algorithms in the structure layer and texture layer dehazing process. Finally, the optimised texture layer and structure layer are recombined to obtain a dehazed image. The experimental results show that the hierarchical optimisation algorithm can preserve the fine texture parts of foggy images, and the edge structure of the scene is clear after dehazing, without distortion in the sky area. The method proposed in this paper is effective.

Inderscience is a global company, a dynamic leading independent journal publisher disseminates the latest research across the broad fields of science, engineering and technology; management, public and business administration; environment, ecological economics and sustainable development; computing, ICT and internet/web services, and related areas.

Inderscience is a global company, a dynamic leading independent journal publisher disseminates the latest research across the broad fields of science, engineering and technology; management, public and business administration; environment, ecological economics and sustainable development; computing, ICT and internet/web services, and related areas.

Inderscience is a global company, a dynamic leading independent journal publisher disseminates the latest research across the broad fields of science, engineering and technology; management, public and business administration; environment, ecological economics and sustainable development; computing, ICT and internet/web services, and related areas.

Inderscience is a global company, a dynamic leading independent journal publisher disseminates the latest research across the broad fields of science, engineering and technology; management, public and business administration; environment, ecological economics and sustainable development; computing, ICT and internet/web services, and related areas.

Fault diagnosis of power distribution networks is a significant task for power system operation and maintenance, which can support rapid and accurate identification, localisation and repair of various faults that may occur in power distribution networks. To improve the accuracy and efficiency of fault diagnosis, this paper proposes a deep adaptive learning model for dealing with the data-driven identification problems in power systems. In the deep adaptive learning model, an adaptive shortcut learning scheme is designed to adaptively adjust the aggregation between the convolutional layers and the attention modules. In this manner, the hidden features can be effectively captured by adaptive learning with the proposed scheme. Therefore, the proposed model can effectively make online decisions on the fault types in high-voltage AC and DC test transmission. The experimental case study and comparison study in this paper demonstrate the reliable performance of the proposed model for online fault diagnosis of power distribution networks, which also show its capability and feasibility for online implementation.