A two-stage dynamic optimization strategy for wind-thermal-energy storage systems in energy and frequency regulation ancillary service markets with adaptive opportunity cost quantification

Abstract

Against the backdrop of high-penetration renewable energy integration, wind-thermal-storage integrated systems need to collaboratively participate in energy and frequency regulation ancillary service markets to enhance overall revenue. However, traditional optimization models neglect the opportunity costs among wind, thermal, and storage resources across multiple markets, leading to rigid resource allocation and profit losses. This study proposes a dynamic opportunity cost quantification method and a two-stage optimization framework combining Markov Decision Process (MDP) and Receding Horizon Control (RHC). The first stage uses MDP-based value iteration to generate day-ahead optimal dispatch strategies, while the second stage employs RHC for real-time rolling adjustment of unit outputs and market bids with adaptive opportunity cost weights. Simulation results demonstrate that the proposed model increases the system's total net revenue by 28.7 % compared to traditional approaches without considering opportunity costs and by 18.9 % versus static opportunity cost models. The framework innovatively enables flexible multi-time scale decision coordination, dynamic opportunity cost weighting, and reduced computational complexity via hierarchical optimization, providing efficient and adaptive decision support for multi-energy systems in electricity markets.