To achieve high-efficiency consumption of new energy, it is imperative to "build a new power system with new energy as the main body". In response to this problem, this paper proposes a multi-time scale rolling coordinated optimal dispatch model of CSP-PV-W that takes into account the maximization of new energy consumption. Firstly, an electric heating device is introduced from the source side to convert excess wind power and PV into heat energy and store it in the TES of the CSP plant; Secondly, according to the characteristics that the prediction errors of wind power and PV become smaller with the approach of operation time point, the multi-time scale coordinated optimal scheduling is introduced. Establish a day-ahead dispatching model based on wind, PV and load forecast curves, combined with thermal power and CSP plants; Then, the day-ahead dispatch result is used as an input variable for intra-day rolling, combined with ultra-short-term wind power, PV and load forecast data, an intra-day dispatch model is formulated, and then the rapid response characteristics of CSP plants are used to balance wind power and PV forecast errors with thermal power plants; Finally, the feasibility of the method are verified by the actual grid architecture of the Jiuquan New Energy Base.