• Advisors – Gauthier Picard
• Contact – send application to Gauthier.Picard@emse.fr
• Location – Laboratoire Hubert Curien & Institut Henri Fayol, Mines Saint-Etienne
• Team – Connected Intelligence & Computer science and Intelligent Systems Dpt

Summary

In this work, we wish to investigate the concept of decentralized markets in microgrids, as to improve local load balancing as much as possible within a distribution network. The idea is to use multiagent and distributed optimization techniques to implement this market in a decentralized manner (without requiring the DSO to enforce the load balancing). Previous works by Cerquides et al. [1,2] proposed the EAP formalisation and some algorithms to clear the market with both discrete and continuous offers [1]. However, there still exists tracks to follow:

• How to use the schema in a repeated market?
• How to define the offers made by agents, especially when repeating the clearing phase?

The first question means that the current model only consider one turn of market clearing, while it could be possible to revise offers and then ask for another market clearing at high frequency many time a day. The second one should answer the issue of how to fix price with respect to the context (common market price, weather, past transactions, etc.).

References

• [1] Cerquides, J., Picard, G., and Rodríguez-Aguilar, J. (2015d). Designing a marketplace for the trading and distribution of energy in the smart grid. In 14th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), pages 1285-1293. International Foundation for Autonomous Agents and Multiagent Systems.
• [2] Cerquides, J., Picard, G., and Rodríguez-Aguilar, J. (2015b). Defining a continuous marketplace for the trading and distribution of energy in the smart grid. In Second Workshop on Interfaces between Multiagent Systems, Machine Learning and Complex Systems, pages 37-48.