For virtual power plants(VPP) composed of wind power generation(WPG) and electric vehicles(EV), uncertainty of wind power output, deviation between wind power forecast and actual output, randomness of EV and the volatility of market prices need to be considered when participating in market bidding. Considering above factors, two stage bidding model for VPP to participate day-ahead and real-time market is proposed. VPP can not only reduce the bidding deviation of wind power through EV charging and discharging, but also make bidding strategies according to the price signal to realize peak clipping and valley filling. Based on cooperative game theory, the Shapley method is used to reasonably distribute the total profit between WPG and EV according to their respective contributions. Finally, feasibility and effectiveness of proposed model are verified by an example. The results show that VPP can increase the profit and reduce the risk caused by output and price uncertainty, and provide reference for the construction of new energy participation in the spot market.