In recent years the amount of renewable energy sources has been massively increased. The power grids have to face great challenges, e.g. supply fluctuations caused by weather changes or efficient control of consumers. Therefore, power grids have to adapt their energy demand to the available energy sources in order to ensure operability. Smart grids and smart metering promise to overcome these challenges. However, in order to control the demand of a grid, it has to be monitored as accurately as possible to gain good results. This creates a conflict between Quality of Information (QoI) and privacy because a greater amount and accuracy of data results in a decrease of privacy.
In this paper we present a privacy-friendly distributed algorithm for the energy management in smart grids that is able to shift and adapt loads and use smart encryption of user data. The algorithm uses a bucket principle where each participant has an own bucket that is encrypted individually and therefore only accessible by a trusted server and himself. In order to transfer additional information needed by the algorithm counters are used that are concealed by random values thus making it impossible to trace the source of the information. By combining all privacy measures, a high level of privacy can be achieved, from the second round of the algorithm on even complete privacy. Results show that the gain in privacy is borne by a data volume overhead.