Exemplifying Conflict Resolution in Multi-objective Smart Micro-Grids


Distributed autonomic management systems following contradictory objectives raise difficult design challenges [2]. We proposed a generic architecture to address this concern and exemplified it via manager integration solutions for multiobjective micro-grids (low-tension networks of the size of a district [4]) [3]. This demo showcases some of these sample implementations via the MisTiGriD simulation platform with the aim of inspiring designers facing similar challenges. Power management in electrical grids, i.e. balancing productions and consumptions, is a primary concern. At the same time, in residential micro-grids, end-user objectives such as comfort or domestic usages directly conflict with power control goals. For instance, heating during cold winter nights would generally conflict with a consumption limitation goal. A viable approach for handling conflicts in autonomic systems consists in introducing conflict resolution facilities in the management layer [2]. Such solutions allow heterogeneous managers to integrate with each other by negotiating conflicting goals. In the case of micro-grids, integration relies on power management organisations [3] where usage goals and power management goals are negotiated via variable organisational patterns [1], [3]. The resulting system behaviour may sacrifice some lower-priority objectives in order to fulfil more important ones, or find a good-enough trade-of, depending on changing environmental conditions, variable manageable resources and dynamic user specifications. MisTiGriD was conceived in order to showcase such conflict resolution mechanisms. This demonstrator is based on a simulation of a district micro-grid, providing the experimenter with partial control such as switching appliances on and off or controlling atmosphere temperature. MisTiGriD also allows the experimenter to set variable, conflicting goals to simulated autonomic systems, for instance, “comfort” goals for smart heaters or a maximum consumption for the district grid. Thus, the experimenter can run a variety of scenarios and explore the capabilities and the limits of the management system, and in particular the way in which conflicts are solved by managers.

DOI: 10.1109/SASOW.2013.10

Extracted Key Phrases

3 Figures and Tables

Cite this paper

@inproceedings{Frey2013ExemplifyingCR, title={Exemplifying Conflict Resolution in Multi-objective Smart Micro-Grids}, author={Sylvain Frey and Ada Diaconescu and David Menga and Isabelle M. Demeure}, booktitle={SASO Workshops}, year={2013} }