Improving Load Forecast in Prosumer Clusters by Varying Energy Storage Size

Forecasting already plays a crucial role in the current electricity grid as a key part of planning and energy management activities. Its role will be pivotal in the envisioned smart grid, as its stakeholders are expected to interact in realtime. The amount of data to be considered e.g., from the smart meters, will sky-rocket; however consistently accurate forecasting will still be needed in order for the smart grid stakeholders to make informed decisions and transact. Today forecasting efforts concentrate on increasing the accuracy of algorithms, especially by grouping energy producing or consuming entities such as the prosumers. However due to the dynamic nature of the smart grid, errors still remain as a result of dynamic and highly volatile events. The usage of distributed storage (potentially strategically positioned) that may exist in smart cities, such as electric cars, aggregations of residential storages, small scale industrial facilities etc., could be used in order to assist towards achieving better forecasting accuracy for groups of prosumers. The latter would enable them to participate in the new smart grid offerings such as smart city level marketplaces with more confident actions and strategies. We investigate here the role of distributed storage in residential areas, as well as a mean towards creating communities (groups) of prosumers that feature better forecast energy behaviour.