Abstract
In the last years, Norwegian distribution grid com-panies (DSOs) have seen a dramatic increase in the demand for electrical energy. Several customer connection requests are put on a waiting list until more grid has been built, which may take years. Therefore, it is crucial to optimize the utilization of the existing distribution grid. The characteristic behavior of both existing and new loads determine whether there is capacity in the grid to connect new grid customers. Data describing the existing loads are available from smart meter data, but is currently not used for the purpose of grid planning or connection requests. This paper describes a novel methodology for utilizing smart meter data for modelling the load of new customers, and compares its performance to traditional load modelling methods used in grid planning today.