Reliability Assessment of Main Equipment in the Distribution Grid

Distribution network is an indispensable part of the power system, which covers a wide area and has a great impact on the reliability of system operation. It is of great significance to conduct an in-depth study on the reliability of distribution network power supply in order to obtain a faster and more accurate analysis method [1]. The reliability analysis methods of distribution network power supply mainly include simulation method (Monte Carlo method), analytical method (state enumeration method, fault mode impact analysis method and minimum path method) and artificial intelligence method (neural network algorithm). In recent years, Yingxiang Wang et al. [2] proposed a distribution network reliability assessment method based on Improved-Elman (IElman) feedback-type dynamic neural network to measure the degree of influence of historical information on the future state, and optimized the relevant parameters through the thoughtevolutionary algorithm to improve the accuracy of neural network assessment of complex problems. Ouyang Sen et al. [3] Establish a distribution network reliability assessment index system from four dimensions: power supply continuity, LRP impact, global and comparative, and provide a refined description of the outage impact on users. Gu Jiahao et al [4]. In order to realize the real-time dynamic assessment of distribution network reliability, a distribution network reliability assessment method based on equipment health degree evaluation model is proposed based on the health index theory. Changkai Shi et al. [5] establish a quantitative relationship between the distribution network equipment health index H and the real-time equipment failure rate; the load point failure rate and load point failure time are used to quantitatively evaluate the overall risk of the distribution network. Currently, there is no model that can perform real-time dynamic assessment of distribution network reliability, so the establishment of a distribution network reliability assessment model with spatiotemporal discrimination can be considered.