Impact Statement—Patrol robots play essential roles to ensure safety of the distribution networks. The knowledge acquisition capability in this complex scenario is key for patrol robots to make decisions. This capability can be realized by the knowledge graph embedding methods. When the accuracy rate of one embedding method is high, it can be put into practical work. The proposed method adds the strongly associated entities into the information aggregation process of an entity. Consequently, the accuracy rate of the proposed method is about 10% or more higher than those of the other methods. The accuracy rate for the simple task can be larger than 95%. For complex tasks, the proposed method could also provide high relative accuracy. With the proposed method, patrol robots can be applied to some analysis work in the field, and the result could have interpretability because of the graph structure.

With the popularization of smart grids, patrol robots have become critical devices in the distribution networks to check the state of equipment. In order to enrich the knowledge of patrol robots in this complex scenario, this article presents the graph-based knowledge acquisition method with convolutional networks for distribution network patrol robots. The proposed method uses a graph convolutional network-based path-related embedding algorithm to complete the knowledge of the distribution network knowledge graph. The proposed algorithm generates the embeddings of entities and relations through aggregating the associated entities in the associated paths, instead of only the connected entities. The graph convolutional network consists of multiple graph convolution layers, and the message-passing process treats different entities discriminatorily according to the association strengths. For determining the plausibility of the knowledge, a scoring function is provided with the convolution...