Abstract:

In order to improve the economy and reliability of a high proportion of new energy power system, a joint optimal dispatch strategy for units based on master-slave game considering the uncertainty of wind and solar power is proposed. Firstly, the transformer long short-term memory(LSTM)hybrid model is constructed to predict the wind and solar power output, and the Wasserstein distance is used to establish the dynamic confidence interval to describe the uncertainty of wind and solar power output. Secondly, a two-level game model between wind and solar power generators and thermal power units is established, and the risk premium and the penalty mechanism of wind and solar abandonment are introduced to optimize the output plan and reserve capacity. Finally, a case study is conducted. The results show that the proposed strategy can effectively reduce the wind and solar curtailment rate and improve the operation economy of the system, which provides a solution for the uncertain optimal dispatch of high proportion of new energy power systems.

Key words: uncertainty, master-slave game, Transformer-LSTM hybrid prediction, dynamic confidence interval, power dispatch