Because of the instability of photovoltaic output, energy storage equipment is required in an area containing a high proportion of photovoltaic output to achieve the balance of energy supply and demand. In order to maximize long-term economic benefits, an optimization method of distributed energy storage with high proportion of photovoltaic output area should be analyzed, and the site selection and capacity allocation scheme of distributed energy storage should be designed. First, in view of the intermittence of photovoltaic output, an algorithm for selecting characteristic photovoltaic curves is established based on an improved rapid sorting algorithm. Secondly, the selection algorithm of the energy curve is designed using an FCM fuzzy clustering algorithm. The selected curve is analyzed to get the characteristic electric quantity of each user. Based on the characteristic power and cost, the position selection and capacity equation of the energy storage station are obtained. Thirdly, the basic idea of a greedy algorithm is adopted to solve the equation, and the optimal energy storage configuration equation is obtained. By optimizing the equation, the energy storage capacity can be allocated to different users. In addition, the influence of future factors on the energy storage configuration is also considered. Finally, in order to verify the effectiveness of the algorithm, the experimental analysis is carried out based on the data of an industrial park, and good results are obtained. This work is supported by the National Natural Science Foundation of China (No. 51877044).