Abstract: The large-scale integration of renewable energy into the grid greatly increases the peak shaving burden of thermal power plants. However, the economy and safety of the thermal power units cannot be guaranteed in the process of deep peak shaving, so they lack the enthusiasm to actively participate in the peak shaving. In order to solve this problem, this paper first calculates the costs and benefits of the thermal power, the wind power, and the photovoltaic power plants in the peak shaving transaction process, and builds an optimal scheduling model for the wind, photovoltaic, hydropower, thermal power and energy storage combined system taking into account of the constraints of peak regulation initiative. Secondly, considering the forecast error of wind power and photovoltaic power, the system reserve constraints are set based on the method of chance constrained programming. A sampling-based method is adopted to transform the opportunity constraints into the deterministic constraints, realizing the rapid solution of the model. Finally, simulation analysis is carried out based on the modified IEEE 30-bus system. The results show that the method proposed in this paper can effectively promote the consumption of new energy and improve the economic efficiency of system operation.