In the independent operation of microgrid with high proportion of renewable energy, in view of uncertainties and fluctuations of its output and load power, and waste of renewable energy, it is necessary to optimize the allocation of energy storage capacity so that the operation of the whole microgrid can achieve the minimum pollution emissions and the highest economy. In the optimization planning of energy storage capacity, a two-level optimization model is adopted. The outer optimization model is responsible for solving the investment problem of energy storage capacity planning of microgrid system. The inner model considers the operation cost, energy storage life loss and pollution gas emission during operation. Among them, the rainflow counting method can be used to calculate the life of energy storage equipment, which can fully consider the factors affecting the life of energy storage equipment. Then, the particle swarm optimization algorithm is used to optimize the energy storage capacity of the microgrid, and the results verify the feasibility of this method. Finally, comparing the two calculation methods of fixed life and exact life, as well as the two optimization methods of minimum emission mode and life optimization mode, it provides effective suggestions for the allocation and operation mode of energy storage capacity.