Real-time Regional Auxiliary Service Market Trading Strategies under High Renewable Energy Penetration

The development of regional auxiliary service market (RASM) has allowed neighboring provinces in the same area to provide each other with supports in terms of peak-shaving and reserve capacities, which could greatly promote the system flexibility. This paper introduces the practice of RASM in northwestern region of China and proposes a data-driven trading strategies model for provincial electricity companies who act as the RASM participants. The objective of this model is to maximize the renewable energy utilization rate while reducing the purchase cost of peak-shaving capacities by rolling optimization of the offers/bids and prices in real-time scale. A human experience embedded proximal policy algorithm is adopted to solve the model. Numerical case studies adapted from real-world data proves the advantage of the proposed method and shows the potential of RASM in promoting the integration of renewable energy.