@article{HUANG2026, 
author = {Zheng HUANG and Yi YANG and Wei WU and Laijun CHEN and Hanchen LIU and Sen CUI and Shijie LI},
title = {Configuration Optimization Method for Underwater Compressed Air Energy Storage Based on Distributionally Robust Chance Constraints},
year = {2026},
journal = {Distributed Energy},
volume = {11},
number = {2},
pages = {1-10},
keywords = {underwater compressed air energy storage (UW-CAES), configuration optimization, pipeline pressure loss, distributionally robust chance constraints (DRCC)},
url = {https://www.sciopen.com/article/10.16513/j.2096-2185.DE.25100136},
doi = {10.16513/j.2096-2185.DE.25100136},
abstract = {Underwater compressed air energy storage (UW-CAES), which utilizes flexible underwater air bags to enable constant-pressure charge and discharge, has emerged as a compelling solution for renewable energy accommodation. However, there remains a distinct lack of research focused on parameter optimization to simultaneously reduce the capital costs of UW-CAES and enhance the operational economics of the plant. To address this critical gap, this paper proposes an optimal configuration method for UW-CAES based on distributionally robust chance constraints (DRCC). First, a comprehensive UW-CAES system model is established, explicitly accounting for the impact of pipeline pressure losses on system dynamics. Subsequently, an optimal configuration framework incorporating these pressure losses is formulated to optimize key system parameters, with the dual objectives of minimizing investment costs and maximizing operational revenues. Furthermore, the DRCC approach is employed to reformulate the stochastic chance constraints into tractable linear constraints. This mathematical transformation not only ensures computational efficiency but also facilitates a flexible trade-off between economic optimality and robustness. Case studies demonstrate the efficacy of the proposed methodology: the optimized system maintains a rated discharge power of 60 MW while reducing the required rated charge power to 53.2 MW—an 8.75% decrease compared to the original baseline—thereby significantly improving overall system efficiency. Finally, sensitivity analyses reveal that systematically calibrating the confidence level and Wasserstein radius within the DRCC framework effectively navigates the equilibrium between economic performance and system conservatism.}
}