Abstract: The increasing penetration rate of distributed new energy in the distribution network and the increasing demand for real-time monitoring often result in delays,packet loss,and other phenomena in the centralized transmission of a large amount of measurement data,leading to a decrease in the accuracy of distribution network state perception and even unobservability. In response to the above issues,a state awareness method based on shared fuzzy equivalence relationships is proposes. Firstly,the collected historical measurement dataset of multi-time sections is used as the source dataset,and the real-time collected data with high measurement packet loss rate is used as the target dataset. Write two sets of datasets in the form of fuzzy sets using triangular membership functions. Then,a distance measurement method between vectors is used to calculate the fuzzy equivalence relationship matrices of two sets of fuzzy sets.On this basis,a clustering method using shared fuzzy equivalence relationships is adopted to transfer knowledge from the source dataset to the target dataset,reducing the difference in feature distribution between the two sets of datasets. Finally,a deep neural network is used to label the data corresponding to the high measurement packet loss rate in the low dimensional feature space after knowledge transfer,and the real-time status of the distribution network is obtained. Through simulation tests on IEEE standard examples and distribution network examples of a certain actual city company,the results show that the proposed distribution network state estimation method based on shared fuzzy equivalence relationship can obtain accurate real-time operation status of the distribution network under high measurement packet loss rates,and achieve real-time monitoring of the distribution network under different levels of measurement data loss.