Optimal capacity configuration of wind-photovoltaic-storage hybrid system: A study based on multi-objective optimization and sparrow search algorithm

The deployment of energy storage on the supply side effectively addresses the challenge posed by the intermittency and fluctuation of renewable energy. Optimizing capacity configuration is vital for maximizing the efficiency of wind/photovoltaic/storage hybrid power generation systems. Firstly, a deep learning-based Wasserstein GAN-gradient penalty (WGAN-GP) model is employed to generate 9 representative wind and solar power output scenarios. Subsequently, an optimization model for capacity configuration in the hybrid system is formulated, aiming to minimize total costs and optimize integrated parameter. The sparrow search algorithm is utilized to solve this model. A case study is conducted on a large-scale hybrid system in a northwestern region in China. Based on model calculations, the proposed energy storage allocation across different scenarios can reduce renewable energy curtailment by 3.6 % to 14.7 % compared to the absence of energy storage. Additionally, utilizing time-of-use electricity prices, this solution can yield annual savings of up to 9.158×107 CNY. In comparison to the current local energy storage configuration schemes, the curtailment rate of renewable energy decreases by 0.7 % to 6.2 % in different scenarios. It is worth mentioning that, in most scenarios, the annual average economic benefits from reducing curtailment according to the proposed method are in the same order of magnitude as the increased investment due to energy storage capacities. These findings validate the effectiveness and practicality of the proposed model and solution approach, providing valuable insights for planning wind-photovoltaic-storage systems.