<p>To address the uncertainty challenges posed by the high penetration of renewable energy integration, this paper studies the multi-agent optimal trading strategy for independent energy storage power plants participating in the electricity spot market. Based on the day-ahead and real-time linked power market framework, firstly, a Monte Carlo simulation combined with scenario reduction method is employed to construct the uncertainty sets for renewable energy output and electricity prices, and a day-ahead market stochastic programming model is established with the goal of maximizing expected profits. Secondly, aiming at the high volatility of the real-time market, a multi-time window rolling robust optimization mechanism is designed to balance economy and risk through dynamically adjusting charging and discharging strategies. Based on non-cooperative game theory, a multi-round bidding game framework is proposed to solve the Nash equilibrium solutions for each agent, achieving win–win market benefits. Simulation results show that the proposed strategy can effectively reduce deviation power and mitigate wind and solar curtailment, while considering the robustness coefficient to balance risk prevention and economic benefits. Meanwhile, under the day-ahead and real-time linkage mechanism, the state of charge of energy storage remains more stable, and the charging/discharging strategies are highly consistent with load fluctuations, verifying the effectiveness of the proposed model in promoting renewable energy integration, optimizing resource allocation, and ensuring stable market operation. This study provides theoretical support and decision-making references for energy storage participation in multi-time scale electricity market trading.</p>