Abstract: To address the renewable energy accommodation issue in China's power grid and achieve the goal of sustainable development, this paper proposes a bi-level energy storage configuration optimization model considering power quality classification of wind power and photovoltaic (PV) power. Firstly, a power quality factor model is developed by analyzing the prediction errors based on historical actual generation power and forecast data of new energy power stations, and taking into account the relative fluctuations between typical daily loads and renewable output characteristics. On this basis, a bi-level optimization model is formulated with the net revenue obtained from system-configured energy storage as objective function: The upper level optimizes the energy capacity and power configuration based on the investment and operation cost of energy storage and accommodation returns; the lower level generates day-ahead and zonal real-time dispatch plans with the objective of minimizing the system operation cost and balancing cost alongside real-time dispatch efficiency. This bi-level model realizes the optimal allocation of energy storage resources and the collaborative design of dispatch strategies by coordinating the economy of the planning and operation stages. Finally, a multi-scenario probability weighting based on the quality factor is adopted to improve the evaluation framework of the multi-timescale energy storage configuration model, and case studies verify the effectiveness of the proposed model in improving system economy and new energy accommodation capability.