Research on SF6 Leakage Detection Based on Artificial Intelligence Gas Cloud Leakage Algorithm

Sulfur hexafluoride (SF6) gas is extensively used in high-voltage equipment such as transformers and circuit breakers due to its excellent insulation and arc-extinguishing properties. However, SF6 gas leakage poses significant safety and environmental risks. This study aims to develop an AI-based gas cloud leakage detection algorithm that utilizes infrared imaging data and deep neural networks for real-time detection and localization of SF6 gas leaks. The proposed method integrates infrared imaging technology with advanced deep learning algorithms to enhance detection accuracy and sensitivity, reducing the need for manual intervention. Experimental results show that the system achieves an overall detection accuracy of $82.71 \%$, with a missed detection rate of $\mathbf{1 3. 8 8 \%}$ and a false alarm rate of $3.41 \%$. This approach ensures the safety of equipment and personnel, and mitigates environmental impact, providing a reliable and efficient solution for SF6 leakage detection in the power industry.