Abstract: Characterized by volatility and randomness, wind-solar power generation can affect power system stability and safety after its large-scale integration into power grid. We established a model of wind-solar-thermal power storage system considering deep peak load changing of thermal power generating unit and energy storage life collapse cost and proposed load tracking and power smoothing output modes. Taking each subsystem output of wind-solar-thermal power storage as the decision variable, we established an optimal scheduling model aiming at the lowest operation cost, the most dissipation of new energy, and the optimal environmental protection. And we solved the dispatching model of NSGA-Ⅱ algorithm to obtain pareto solution set. Finally, we used AHP decision method to determine the optimal dispatching scheme, realized output tracking loading and energy system power smoothing of energy system, and reduced the effect caused by the integration. The simulation results show that the scheduling strategy under two modes can improve the economy of wind-solar-thermal power storage system and the new energy dissipation capacity, reduce polluted gases emission, which verifies the effectiveness of the optimal scheduling model.