The ability of power grids to quickly recover from extreme environmental disturbances such as natural disasters plays a pivotal role in social stability and national security. In This paper, optimizing the resilience of new energy-penetrated power grids was investigated by integrating recovery methods of different timescales, such as emergency dispatching and gradual equipment repair. First, considering the fluctuation of new energy outputs, a short-term optimal scheduling model of new energy-penetrated power grids on a given topology was established. Second, in order to solve the problem that the topology formed by connecting all undamaged devices after the disaster was not necessarily the optimal topology, the genetic algorithm was used to reconstruct the topology of the power grid after the disaster. Finally, considering the sequential repair process of damaged equipment, the greedy algorithm was used to generate the repair sequence, that was, the power equipment whose repair leads to the largest benefit increment was preferentially selected each time. Simulation results show that the integrated recovery scheme can effectively improve the overall power supply benefit of the whole restoration process of new energy-penetrated power grids.