Abstract: With the number of electric vehicles (EVs) increasing, the time-delay problem becomes more and more obvious when they participate in the real-time optimal scheduling of the grid demand response after they are connected to the grid. This problem not only affects the quality of power grid response, but also reduces the response income of the users. Based on this background, a real-time demand response scheduling strategy for EVs based on the cloud-side collaboration is proposed in this paper. Firstly, the EV schedulable load model is analyzed; Secondly, the cloud edge collaborative task unloading strategy is established based on the EV mobility characteristics; Finally, the scheduling decision model is established with the minimum load difference of the central cloud and the maximum response benefit of the edge cloud as the objectives. In order to analyze the response revenue of the EV users with time delay, this paper further illustrates the users' benefit from the user response quantity and the user response waiting time. The simulation results not only verify that the proposed method can effectively reduce the load peak-valley difference and improve the revenue of aggregator and user side, but also prove its superiority in the process of large-scale EV participating in real-time scheduling.