A method for state estimation of distribution network based on the classification of distribution transformer load

Abstract

© 2016 IEEE. Based on the data classification of distribution transformer history load, a distribution network state estimation methods has been presented on this paper. By using the K-mean clustering technology to classify historical load data, the validity and accuracy of the load model and data have been promoted, and the error rate of the state estimation of distribution network is reduced. The problem of the insufficient field measured data is solved by using the state estimation (SE) algorithm, and compensated by using the historical data record, pseudo and virtual measurement values, in order to obtain operation necessary for a consistent trend with minimum input data set. It is proved that the algorithm can meet the demand of practical application through the field example.