Abstract: Large-scale utilization of new energy such as wind and solar is an important way to achieve the goal of dual carbon, but its uncertainties affect the operation stability of power systems and the capability of absorbing new energy. This paper develops a capacity optimization allocation method considering uncertainties for wind-solar pumped storage power generation systems. First, a random distribution function is used to describe the output characteristics and load distributions of wind power and photovoltaic power generation, and a mathematical model of complementary power generation system is developed. Then, we construct a two-layer capacity optimization configuration model based on the information gap decision theory and the entropy weight method. Finally, we use a multiverse optimization algorithm to solve for capacity optimization configuration schemes. Calculations show that investors can adopt different allocation strategies for source-load uncertainty according to their investment intentions, and the operating results obtained can well meet the requirements in different operation modes and multiple operation scenarios. Our method can help reduce the comprehensive cost significantly, improve the capacity of new energy consumption, and maintain the long-term operation of pumped storage power stations.

Key words: source load uncertainties, information gap decision theory, pumped storage, capacity optimization configuration, two-tier optimization