Abstract– As distributed renewable generation sources continue to be integrated into the power grid, the concept of Microgrids has gained traction. Here, the integration challenge is not just in the control of an individual Microgrid but also in their coordination. This short communication presents a novel approach to inter-Microgrid coordination and control based upon Multi-Agent systems. An architecture based upon physical agents is presented and is implemented on a dual platform of JADE (environment for developing agents) and Matlab- Matpower (power system analysis tool). The implementation is demonstrated for a reconfiguration scenario involving the preservation of vital loads. The work presents many opportunities for future developments in the domain of resilient self-healing power grids. I. INTRODUCTION The motivation of this project is the place of Microgrids in the future of the electricity grid. In [1], it is realized that Distribution Systems increasingly coming to resemble Transmission Systems, both trending towards real-time controlled Smart Grid and with the introduction of Variable Energy Resources. Thus, Distribution Systems will form temporary semi-autonomous Microgrids, as shown in Figure 1. The Microgrids benefits are congestion relief, postponement of new generation or delivery capacity, response to load changes, local voltage support, promote high penetration of renewable sources, dynamic islanding, and improved generation efficiencies through the use of waste heat [2], [3], [4]. Due to these benefits the power grid will deal with systems with multiple microgrids. It is required advanced modeling and the development of control systems working with inter-Microgrid coordination, this short communication proposed the first approach of an innovative technique for that. This research is supported by the MIT-Masdar Collaboration. S. Rivera is with the Mechanical Engineering Department, Massachusetts Institute of Technology, Cambridge, MA, 02139 USA, as Postdoctoral Associate in the MIT Mechatronics Research Laboratory (email: serivera@mit.edu). A. Farid is with the Engineering Systems and Management Department, Masdar Institute of Science and Technology, Abu Dhabi, UAE, as Professor, and with MIT as Research Affiliate in the Technology and Development Program (email: afarid@masdar.ac.ae). K. Youcef-Toumi is with the Mechanical Engineering Department, Massachusetts Institute of Technology, Cambridge, MA, 02139 USA, as Professor (email: youcef@mit.edu). Figure 1: Semiautonomous Microgrids. In Section II is presented a review of the academic literature of Coordination and Control of Microgrids, in Section III is presented the proposed MAS approach. A case study, in order to solve the load supply and self-healing problem of multiple Microgrids, is addressed in Section IV and the conclusions are in section V. II. COORDINATION AND CONTROL OF MICROGRIDS The Coordination and Control of a Microgrid needs to ensure [1]: • New microsources can be added to the system without modification of existing equipment. • The Microgrid can connect to or isolate itself from the grid in a rapid and seamless fashion. • Reactive and active power can be independently controlled. • Microgrid can meet the grid’s load dynamics requirements. In order to insure the before, the Microgrids use the 2 following physical control methods given by the injection of power of the microsources into it [2]: 1. The first physical control method is the voltage regulation through droop, where, as the reactive current generated by the microsource becomes more capacitive, the local voltage set point is reduced. 2. The second physical control method is the frequency regulation through droop. When the Microgrid separates from the grid, the voltage phase angles at each microsource in the Microgrid change, resulting in a reduction in local frequency, this frequency reduction is Coordination and Control of Multiple Microgrids Using Multi- Agent Systems Sergio Rivera, Amro Farid, Kamal Youcef-Toumi