Adaptive and Predictive Energy Management Strategy for Online Optimal Power Dispatch from VPPs with Renewable Energy and Energy Storage

Abstract

Virtual power plant (VPP) aggregates heterogeneous distributed energy resources through a cloud-based access control system providing efficient centralized management, visibility and control. Most of the operation strategies for VPPs are designed based on the day-ahead profiles. However, prediction errors of the renewable energy sources and the load demands can lead to a sub-optimal operation in the dispatch scheduling. In this paper, an adaptive and predictive energy management strategy for an online optimal operation of VPPs is proposed based on the model predictive control technique with a feedback correction to compensate for the prediction error. This strategy is divided in to two sections: a) rolling horizon optimization; and b) feedback control based error correction. In the rolling horizon optimization, a hybrid prediction algorithm based on the integration of the time series analysis and the Kalman filters is used to forecast the output powers of the renewable energy sources and the load demands. The rolling horizon optimization model is implemented as a mixed-integer linear program (MILP) to schedule the operation in accordance with the latest forecast information. The power dispatch schedule is then adjusted based on ultra-short-term error prediction. The feedback control-based error correction is applied to minimize the adjustments for compensating the prediction error and is implemented as a linear program (LP). The proposed strategy is implemented on a VPP in a practical distribution system in New South Wales (NSW), Australia. The simulation results demonstrate the effectiveness of the proposed strategy with better tracking of the actual available resources and a minimal mismatch between demand and supply.