Generation of realistic Smart Meter Data from Prosumers for future energy system scenarios

Future energy systems with high proportion of intermittent and distributed renewable generation need the coupling of the energy sectors electricity, gas and heat into an integrated energy system. In order to achieve supply safety for this novel system, advanced operational concepts will be required. These advanced algorithms, like for example integrated grid state identification and prognosis, require a high amount of data with a high temporal resolution for testing and evaluation. In German electrical energy systems, grid operators rely on the currently ongoing smart meter rollout for the purpose of data acquisition. Nevertheless, these large amounts of data are hard to obtain and are often restricted due to reasons of privacy. Furthermore, this data belongs to the actual grid situation and is therefore not identical to the data expected in future grid scenarios. In this paper, an approach to synthetically generate realistic future smart meter data is proposed. The household technologies and smart meter are modeled with the open-source TransiEnt Library for dynamic modeling of integrated energy systems. Furthermore, the models are tested and evaluated with real smart meter measurement data from different households in Lower Saxony.