An Intelligent Fault Diagnosis Method for Street Lamps

With the continuous advancement of urbanization, higher and higher requirements are put forward for infrastructure construction. Accurate fault diagnosis and timely maintenance of street lamps have become an important part of the lighting system. This paper proposes an intelligent fault diagnosis method for street lamps based illumination detection, narrowband Internet of Things (NB-IoT) technology and machine learning. In this proposed method, the narrowband IoT modules embedded with the illuminance sensor are installed and numbered at the street lamps and the collected illuminance data are uploaded to the server to build the illuminance database. Then the machine learning method is used to learn the modes of the collected data sequences and the street lamp turn-on models are constructed. Further, the real-time illuminance data sequence is processed to realize the fault diagnosis and to judge the fault type of a single street lamp or lamp group, and then feedback to the maintenance staff. To verify the proposed method, one example is also given. The proposed method provides one effective way for fault diagnosis of street lamps.