Optimal Expansion Planning of Microgrids Clusters: A Robust Collaborative Approach

Abstract

Gathering electrical demands and distributed generators into microgrids help to coordinate them. When different microgrids are located at the same geographical area and connected to the same transmission network, they could be self-organized in microgrids clusters. Thereby, microgrids could import/export surplus energy from other microgrids in the cluster. Intuitively, when microgrids engage in clusters, peer-to-peer sharing needs to be considered when planning investment in new assets. This paper focuses on this issue, developing a novel methodology for optimal expansion planning of microgrids clusters. The new proposal ensures privacy of each microgrid, which only exchanges boundary information with other peers. The new optimization model casts as a three-level methodology, accounting for the inherent risk of uncertain renewable generation and demand. To this end, uncertainties are modelled using a polyhedral set, whose bounds are determined using a novel clustering strategy. The resulting bi-level model is solved in a using a tailored algorithm, which embodies robust optimization with Column-and-Constraint-Generation algorithm. The new solution methodology is tested on a benchmark three-microgrids cluster. The results show the capabilities of the new proposal to deal with uncertainties in a robust manner. Moreover, its adaptive feature is illustrated, allowing to tune the level of risk assumed as well as the available monetary budget. Finally, some sensitivity analysis are carried out regarding the fuel cost and the number of microgrids in the cluster, showing up the new solution approach scales well with the size of the system.

Keywords: Column-and-Constraint-Generation algorithm, Polyhedral uncertainty set, Microgrids cluster, Robust optimization