In order to ensure the safety, stability and economic operation of the wind power systems, a rotating reserve capacity optimization scheduling model with the minimum total operating cost is constructed.To improve the speed of solving the model, an improved simulated plant growth algorithm is proposed to optimize the model. In order to solve the problems such as low optimization efficiency and ease of falling into local optimum when solving the model, the reverse learning idea is introduced into the plant growth algorithm, the growth point is inversely mutated to expand the search space of the algorithm, and the intelligent variable step search and the variation mechanism of the elite set are adopted to ensure the fast optimization and improve the solution accuracy. The standard test function is used to verify that the improved algorithm has faster calculation speed and better optimization ability. Finally,the example verification is carried out on the IEEE 30 bus system. The experimental results show that the proposed model can effectively solve the rotating reserve capacity optimization scheduling problem in the wind power systems.