Abstract: Extreme disasters will have a certain impact on the distribution network, and even disrupt the safe and stable operation of the power grid. A large number of data about distribution network equipment and operation are generated during the disaster events, which are large in information contents and high in volatility. In order to timely grasp the damage information of distribution network equipment under extreme disaster events and effectively assist the post-disaster repairing work of the distribution network, a spatiotemporal grid based distribution network multi-source data fusion method is proposed. Firstly, according to the spatiotemporal grid division method, the area where the distribution network is located is partitioned in time and space, and a spatiotemporal grid coding database of the distribution network is established. Then, the characteristic analysis of the multi-source data of the distribution network under extreme disaster events is carried out to extract key time features and spatial features. By calculating the correlation degree of spatiotemporal features, the spatiotemporal grid and the distribution network multi-source data are associated and stored in the spatiotemporal grid coding database. Further, the grid view of the distribution network damage is constructed according to the fused multi-source data of the distribution network. The results of the case study show that the proposed method is effective and reasonable.

Key words: extreme disasters, distribution network, multi-source data, spatiotemporal grid, data fusion