There is a problem that the actual photovoltaic output is ignored in the optimal allocation of energy storage capacity in a photovoltaic micro-grid. To solve this, a photovoltaic output power correction method considering temperature and light intensity is proposed, and a photovoltaic optimization model is established. Secondly, taking the economic optimal energy storage as the target, and taking account of the need to meet the load demand, optical storage joint modeling is carried out. Considering the influence of load power shortage rate and energy spillover ratio, capacity allocation and economic analysis are conducted for five energy storage batteries under the time-of-use electricity pricing of the grid. Finally, an example is given to verify that the proposed model is more suitable for the regional characteristics and conforms to the actual photovoltaic output power. The configured energy storage results show that NAS costs are highest during the project cycle, VRLA-cap costs are lowest, and the total investment costs of V-redox, LPF, and VRLA-B are similar. This work is supported by the National Natural Science Foundation of China (No. 61871259) and the Project of Xi’an Science and Technology Bureau (No. 2020KJRC003).