Research and Implementation of Intelligent Monitoring System for Power Transmission Line Faults

Power transmission lines, though widely distributed, remain vulnerable to various hazards. While visual monitoring technologies have significantly reduced manual inspection workloads, conventional monitoring devices still exhibit limitations under nocturnal conditions, particularly regarding in detail preservation. These constraints necessitate further improvements to ensure the secure and stable operation of power systems. This paper presents a Transformer-based and HVI-GLARE Fusion Low-Light Image Enhancement network (THG-LLIE) that integrates HVI color space characteristics with generative latent feature retrieval methodology. Comparative experimental analysis demonstrates that our proposed algorithm achieves superior image enhancement results across both quantitative and qualitative metrics relative to state-of-the-art enhancement approaches. Additionally, we have developed a domestically produced AI chip based on our proprietary architecture, enabling implementation of an intelligent AI-based detection system for power transmission line fault monitoring.