Asynchronous event driven distributed energy management using profile steering

Distributed Energy Management methodologies with a scheduling approach based on predictions require means to avoid problems related to prediction errors. Various approaches deal with such prediction errors by applying a different online control mechanism, such as a double-sided auction. However, this results in two separate control mechanisms for the planning phase and the real-time control phase. In this paper, we present a two-phase approach with profile steering based control in both phases. The first phase is synchronous and uses predictions to create a planning. The second phase uses profile steering to schedule individual devices in an event driven and asynchronous manner. Simulation results show that this methodology results in an improved power quality and follows the planning better with a RMSE reduction of up to 34%. In addition, it provides more robustness to failure of connection and improves transparency of its actions to prosumers.