Long-Term Coordinated Planning of Multi-Type Energy Storage Considering Typical Wind-PV-Load Scenarios
Abstract
To address the difficulty of coordinating renewable generation variability, load diversity, and energy storage expansion over a long-term planning horizon, this paper proposes a long-term coordinated planning method for multi-type energy storage considering typical wind-PV-load scenarios. Kernel principal component analysis and agglomerative hierarchical clustering (KPCA-AHC) are first applied to reduce annual wind-PV profiles and extract four representative scenarios, while the corresponding load profiles are synchronously selected to preserve source-load correlations. A 2026-2035 planning model is then formulated for pumped hydro storage (PHS), battery energy storage (BES), and compressed-air energy storage (CAES), with the objective of minimizing discounted system cost under capacity-evolution, operation, reserve, renewable-curtailment, power-export, and power-balance constraints. The provincial case study shows a staged expansion pathway: PHS is expanded in 2028-2031, CAES in 2032-2034, and BES mainly in 2034-2035. By 2035, their capacities reach 6200MW, 7737MW, and 2800 MW, respectively, while all scheduled exports are satisfied. The method provides a practical basis for coordinated long-term storage planning in renewable-rich power systems.