Multi-Time-Scales Energy Management for Grid-On Multi-Layer Microgrids Cluster

Xiaowen Xing,Hongmin Meng,Lili Xie,Peng Li,Sergio Toledo,Yuzhi Zhang,Josep M. Guerrero
DOI: https://doi.org/10.1109/spec.2017.8333656
2017-01-01
Abstract:Multi-layer microgrids cluster (MLMC) has been formed in some certain areas with microgrids (MGs) merging into smart distributed system. It relies on an energy management system (EMS) to manage and schedule the optimal power flow among MGs and public power system in a stable and balanced way. Compared with EMS for single MG, energy management optimization for MLMC is much more complex that the interactions with other MGs require to be considered. In this paper a multi-time-scales EMS is proposed to minimize the total operating cost of MLMC. It consists of two parts: day-ahead static scheduling and real-time dynamic compensation. The day-ahead scheduling is an economic optimization based on the forecasting data within 24 hours. The real-time dynamic compensation is presented to compensate scheduling signal based on the real-time operating information, which is caused by the predictive error. Furthermore, charge-discharge protections of energy storage system (ESS) for each single MG are taken into account. A general-structure MLMG is studied as an example, and this problem is solved by mixed integer linear programming (MILP). The effectiveness of proposed EMS is verified in the MG Research Laboratory in Aalborg University, Denmark. The hardware-in-loop simulation results show the proposed EMS has a good convergence and performance in grid-connected mode.
What problem does this paper attempt to address?