A Machine Learning Method Creating Network Models Based on Measurements

Network models are essential to perform power flow analyses. In this paper a supervised regression method creating simplified network models using measurements is presented. It is an iterative method creating a network model by minimizing the difference between measurements and obtained power flow using measured net-exchanges for each node. The method is tested in a case study for the Nordic Synchronous Area considering each bidding zone as a node. The simplified network model is created using a training set and is validated using various validation methods. The obtained reactances are not correct in absolute terms; however results indicate that the obtained power flows using the created network model are accurate enough for several different applications.