Abstract: Due to the impact of the large-scale renewable energy on the safe and stable operation of power system, the flexible resources such as gas turbine and storage are considered to adjust the volatility of the wind power. A two-stage stochastic scheduling model co-optimizing energy and reserve was established, with constraints describing multi-operating states of conventional thermal power units. Then, the Lagrange multipliers corresponding to the two-stage power balancing constraints in the model were taken as the day-ahead and real-time electricity prices, respectively, and the mechanism of marginal cost from the perspective of model was explored. The influence mechanism of flexible resources on system marginal price was analyzed in depth from two aspects: the lacking degree and exuberant degree of demand for flexible resources. Case study illustrates the effectiveness of flexible resources in dealing with the uncertainty of wind power and restraining peak price. Moreover, the influence of coupling of multi-time scale climbing constraints and coupling of day-ahead reserve and real-time output adjustment on system marginal electricity price was analyzed.