A decentralized framework of unit commitment for future power markets

Participants in future power markets will include many new entities brought by the emerging Smart Grid technologies. Such entities tend to behave autonomously to maximize their own benefits without revealing their detailed information, making it difficult to centrally clear the unit commitment based markets considering the security constraints coupled among entities. This paper presents a two level decentralized framework for solving the day-ahead unit commitment problem. On the lower level, each entity solves its own problem to optimize its benefits under the guidance of locational electricity prices, and sends the results to a bus coordinator. On the upper level, coordinators communicate with their neighbours to update the prices by using a distributed subgradient method. It is proved that with probability 1, the method will lead to a consensus on prices among all buses so that the market is cleared. Illustrative examples are provided to demonstrate the effectiveness of our framework.