For the approximation of low-voltage grid dynamics we present an energy based bottom-up model to create highly diverse household load profiles. In an object-oriented approach, nodal loads are represented by classifications of appliances and distributed to the phases in order to consider the characteristically unbalanced load of residential customers over time. Probability parameters, regarding time of use, corresponding duration of use and the magnitude of power consumption, are derived from a combination of statistical data and global load profiles. These parameters build probability models from cross-sectional data for specific points in time which are processed by a probabilistic load-flow along the time axis.
A simulation of a rural low-voltage grid using detailed information about the topology, renewable energy sources and residential energy consumption is performed. Results are validated and compared to high-resolution measurements of the real grid. The proposed model - embedded in an appropriate simulation scheme - will support the further development of planning, operational and control strategies for today's and future low-voltage grids.