Abstract: Considering the impact of the uncertainty of photovoltaic output and load demand on the planning and operation of the building integrated photovoltaic(BIPV) system, a method of cooperative allocation of photovoltaic and energy storage capacity and energy scheduling based on robust optimization is proposed. Firstly, the uncertainty of PV power generation and building load is modeled, the whole life cycle system costs and benefits are considered, and the optimal allocation of PV energy storage capacity and system economic dispatch model are established according to the characteristics of time-of-use tariff distribution, so as to give full play to the advantages of energy storage “peak-valley arbitrage”. Secondly, based on the robust optimization theory and column-and-constraint generation(C&CG) algorithm, the model with uncertain variables is transformed into the main and sub-problems with uncertain variables and solved alternately. Finally, the simulation analysis proves that the “time inverse distribution” characteristics of photovoltaic power generation and building loads lead to the “worst” scenario. Meanwhile, comparing the proposed model with the traditional mean deterministic model, it is found that the daily operating returns of the robust optimization method are 25% higher than that of the mean model in the “worst” scenario. The proposed model and method can effectively cope with the large fluctuation of power generation and load curve and guarantee the system benefits, and provide a reference for the energy storage construction planning and operation scheduling design of BIPV system.