A knowledge graph-based reliability assessment method for overhead distribution lines

Medium-voltage overhead distribution lines are characterized by large scale, wide distribution, diverse types, and complex operating environments, serving as the key to safe and reliable power supply of distribution networks. Aiming at the difficulties in analyzing operation and maintenance text data and evaluating power supply reliability for medium-voltage overhead distribution lines, this paper constructs a knowledge graph of medium-voltage overhead distribution lines by utilizing multi-source heterogeneous data such as equipment ledgers and operation & maintenance logs. On the basis of the fault mechanism and correlation analysis of individual line equipment, a time-varying fault rate model of equipment and an overall reliability evaluation model for overhead distribution lines based on dynamic Bayesian networks (DBN) are established. Firstly, a knowledge graph for the operation, maintenance and reliability evaluation of medium-voltage overhead distribution lines is established. According to the different structural characteristics of multi-source heterogeneous data, data association rules are formulated to construct the schema layer, and entities are extracted from text data to build the data layer. Secondly, recurring power outage factors are analyzed based on the knowledge graph. The correction coefficients and weights of major outage factors are evaluated and adjusted, and a real-time fault prediction model for overhead distribution equipment considering faults caused by severe meteorological factors is proposed. Then, DBN is adopted to establish the overall reliability evaluation model of overhead distribution lines to simulate the evolution process of equipment faults under multiple operating states. Finally, case verification is carried out using actual distribution network data from a region in Zhejiang Province. The results show that the proposed method can accurately evaluate the operational reliability of overhead distribution lines, identify weak links, and effectively guarantee the safe and stable operation of distribution networks.