Assessment method of distribution network health level based on multivariate information

Abstract

In order to enhance self-monitoring and self-diagnosis capabilities in smart distribution networks, this paper proposes a method for assessing the health level of the network based on multivariate information. First, we construct an evaluation indicator system for the health of the smart distribution network by integrating the smart distribution network information system. Next, we utilize the improved back propagation (BP) neural network and multivariate indicator information to calculate the health indexes of both the grid layer and equipment. We then solve the health index of the equipment layer based on network topology and goal-oriented methodology. Furthermore, by utilizing the health information of both the equipment and grid layer, we apply fuzzy evaluation and Dempster-Shafer (D-S) evidence theory to obtain the health level of the distribution network. We provide a comprehensive evaluation of the overall health status of the smart distribution network. Finally, the proposed method is validated using data from a regional distribution network. The results demonstrate its effectiveness in improving the smart distribution networks’ overall health and stability by enabling more effective self-monitoring and self-diagnosis.