In order to solve the problem of low profit of virtual power plant caused by the uncertainty of wind power, photovoltaic and load demands, an optimal scheduling model of CCHP virtual power plant based on multi-scenario is proposed. The continuous time scenarios are generated by latin hypercube sampling to simulate the wind power and photovoltaic output and the fluctuation of multi-energy load with time. The generated scenarios are reduced by simultaneous backward reduction to obtain typical scenarios. Moreover, a two-stage coordinated optimization model for multi-scenario CCHP virtual power plants is constructed. In the real-time stage, an operational model considering multiple costs and constraints is established, and the proposed model is optimized by particle swarm optimization algorithm to obtain the best multi-scenario output scheme. In the real-time correction phase, the regulatory cost is added to adjust the output of the controllable equipment to eliminate the volatility. The simulation analysis in typical cases verifies that the proposed model can eliminate the influence of uncertainty of wind power, photovoltaic and load, and promote the complementary and coordination of operation for CCHP virtual power plant.