Research and application of intelligent monitoring system for transmission lines with deep learning integration

Résumé fourni par la source

A smart monitoring and photographing system for transmission lines has been proposed. It is integrated with deep learning. The system is used to address the challenges of operation and inspection of high-voltage transmission lines under adverse weather conditions. The aim of this system is to enhance the visual operation and inspection capabilities of transmission lines. These capabilities are needed under complex climatic conditions such as haze. All-weather monitoring of the transmission line status has been achieved. This is done by deploying smart monitoring and photographing devices. A residual-adaptive enhanced AOD network (RAE-AOD) has been designed. It is used for image de-hazing. Image quality has been effectively improved by it. Performance tests have been conducted in real-world environments. The results have shown that high image peak signal-to-noise ratio (PSNR>17 dB) and structural similarity index (SSIM close to 0.9) are achieved by the RAE-AOD network. This happens under light, moderate, and heavy haze conditions. Image clarity and detail representation have been enhanced because of this. Moreover, stable operation and real-time data transmission under extreme climatic conditions are ensured by the design of the smart monitoring and photographing device. Strong technical support is provided by this for the operation and maintenance management of transmission lines.