The aim of this document is to provide an overview of the management and supervision techniques for a micro grid, as well as providing the decisions made concerning the architecture of the micro grid involved in the DER project.
A micro grid is a system that aggregates generator devices (active equipments or sources), consumers (passive equipments or loads), energy storage devices (batteries) and hybrid devices (like EV that is able to both produce or consume energy). All are intelligently coordinated so that they produce electric and thermal energy in a sustainable, safe and efficient way.
The micro grid can be seen as a small smart grid that incorporates smart measurement devices (smart sensors), control methods and communication systems between its internal equipments. The control system of a micro grid must take into account economical criteria (in order to maximize the profit of the micro grid) and technical criteria (to provide safety and efficiency). The energy management system not only should take into account its own benefit, but also should provide support to the public electric distribution network when possible. 

There are three possible operation modes in a micro grid:
  • Normal operating mode, connected to the grid, sets the operation points of its devices to optimize its operation following reliability, economics and micro grid needs criteria.
  • Emergency operating mode, connected to the grid, a balance between the micro grid needs, reliability and energy estate of the grid should be considered
  • Island operating mode, stand-alone, the micro grid should cover its energy needs only with its own energy resources.
In order to provide an optimized control of the micro grid, not only the current values of the inner devices should be taken into account, but also forecasted values and external information, such as energy prices and weather forecasting.
The control system can be designed following a centralize configuration, with a unique agent (head agent) or following a distributed configuration of a Multi-Agent System (MAS) involving several agents acting and being able to take their own decisions. Furthermore, there are hierarquical configurations that merge the two architectures mentioned previously. In such architectures, devices are organized in a hierarchical scheme so that each level reports to the element of its higher level.

Inside project DER, we pretend to implement two different control architectures:
  •  Hierarchical, defining a central control level in charge of the economic control and who sends the setups technical points to the lower level, and several local controllers who manage technically the devices under their control.
  • Distributed, in which every agent considers only its own benefit. There is not any central agent who cares to the global micro grid optimization.