The invention provides a method and device for predicting standby demands of a power system, and relates to the technical field of auxiliary service market standby demand prediction. The method comprises the steps of obtaining thermal power generating unit data, photovoltaic unit data, wind power generating unit data and user load data in a power system; determining a preliminary standby demand ofthe power system according to the thermal power generating unit data, the photovoltaic unit data, the wind turbine generator data and the user load data; obtaining the user type of the power system,determining the power utilization satisfaction of the user according to the preliminary standby demand, and determining the payment satisfaction of different types of users; determining the comprehensive satisfaction of the user according to the power utilization satisfaction of the user and the payment satisfaction of different types of users; and determining a prediction result of the standby demand of the power system according to the preliminary standby demand and the comprehensive satisfaction of the user. According to the invention, a more accurate and objective spare demand prediction method can be realized.