다중 센서 융합 기반 화재 상태감시 및 소화 시스템

초록

Wind turbines have been installed worldwide to produce green and clean energy. Their widespread use can help to decrease carbon emissions, thereby lowering the overall temperature of the earth. Wind turbines, being exposed to severe environments, are prone to several disasters including nacelle fires and blade failures. These can be very serious because the nacelle is the part that houses a generator, gearbox, and other critical mechanical and electrical components. These components are subject to high mechanical and thermal stresses, which may lead to overheating and fires. Existing automatic fire detector systems, such as a flame detector, have several limitations including low availability and delay in detection. To overcome these limitations and thereby ensure reliable operation of wind turbines, this study proposes a novel fire detection and extinguishing system. The study focuses on sensor fusion by combining an infrared camera and LiDAR (Light Detection and Ranging) system to detect location of overheating in the main components of wind turbines. The infrared camera is used to detect overheating and fires, while the LiDAR is used for providing information for environment recognition. Two experiments are performed to verify the effectiveness of the proposed system. In particular, the experiments were conducted to determine the optimal location for the system's ability to localize faults with precision. Fire experiments were also conducted to verify the automatic fire detection performance of the system. The results demonstrate that the proposed system demonstrates excellent performance in detecting fires and overheating, making it a promising solution for fire monitoring of nacelle in a wind turbine. The proposed method can also be effective for fire and overheating detection in many other real-world applications including manufacturing plants, factories, power plants, and other infrastructures because these places require a robust and highly accurate fire detection system.

키워드