Power Flow Calculation Method of Three-phase Unbalanced Distribution Network Based on Data-mechanism Fusion Model

Abstract

A large power error is caused by traditional power flow algorithm applied to actual three-phase unbalanced distribution network, so a novel power flow calculation method is proposed based on data-mechanism fusion model. Through the analysis of the power flow calculation errors, it was found that the main reason for the errors is that the actual transformer loss cannot be truly calculated based on the mechanism model of transformer under the condition of three-phase unbalance. On basis of the historical operation data in the distribution network, a data and mechanism fusion modeling method is adopted, and the mechanism model is used to guide the establishing of the data model. Firstly, grey relational analysis is used to select the key feature variables for the soft sensing model of transformer loss. Meanwhile, combined with the Tent-SSABP algorithm, the soft sensing model of the transformer loss value is established. Then, it is embedded into the three-phase power flow algorithm to replace the mechanism model of transformer, and a data-mechanism fusion power flow calculation has been achieved. Finally, the power flow calculation method proposed is verified in an actual research projection. The results show that the proposed method has higher accuracy than the traditional power flow algorithm using the mechanism model of transformer.