Distributed algorithm for dynamic economic power dispatch with energy storage in smart grids

Abstract

The dynamic economic dispatch problem with energy storage in a smart grid scenario is studied, which aims at minimising the aggregate generation costs over multiple periods on condition that the time-varying demand is met, while physical constraints on generation and storage as well as system spinning reserve requirement are satisfied. In our model, energy storage devices are incorporated for not only inter-temporal energy arbitrage to reduce total generation cost, but also providing spinning reserve to share generators' burden. To solve this problem, we assume that the communication networks are strongly connected directed graphs and propose a fully distributed algorithm based on the `consensus-like' iterative algorithm and the alternating direction method of multipliers. Our algorithm is distributed in the sense that no leader or master nodes are needed, while all the nodes conduct local computation and communicate merely with their neighbours. Numerical simulation is included to show the effectiveness of the proposed algorithm.