Exploring Spatio-temporal patterns of power faults in Guangzhou

Using the power grid failure data of Guangzhou City during 2016-2018, we use the heatmap visualization, Moran’s I method to explore the spatial distribution pattern of power faults and identify the power fault-prone areas at different spatial scales in Guangzhou Power Grid. We also use DBSCAN clustering method in the analysis of the spatiotemporal features to obtain the high occurrence areas in the high-risk time intervals. The results can help the relevant institutes to take preventive measures specific to places and periods while facing power faults.