Multi-timescale scheduling strategy for multi-microgrids with accelerated alternating direction method of multipliers and stochastic model predictive control

Abstract

The superiorities of renewable energy, such as wind and solar energy, have promoted the development of microgrids (MGs) and multi-microgrids (MMGs). However, how to coordinate the scheduling and transactions of MMGs with multi-timescale is still an important issue. This paper presents a scheduling and trading strategy of MMGs with two time-scales: day-ahead and intra-day. In the day-ahead scheduling stage, a MMG system with peer-to-peer connection is considered. Based on the idea of distributed updating parameters and adaptive selecting values in Alternating Direction Method of Multipliers (ADMM), an accelerated ADMM algorithm named improved adaptive accelerated ADMM (IAA-ADMM) is proposed, which is modeled and solved in a distributed manner. In the intra-day scheduling stage, based on the day-ahead scheduling, this paper utilizes stochastic model predictive control (SMPC) to optimize the intra-day model, which helps address the uncertainties of wind, solar, and load forecasting. The effectiveness of the proposed approach is validated using numerical examples. The results show that the IAA-ADMM provides higher stability and faster convergence and facilitates easier implementation. The SMPC shows higher economic performance and has a higher application potential.