Artificial intelligence-empowered modeling and management of flow batteries: A mini-review

Abstract

Flow batteries are pivotal for grid-scale renewable energy storage due to their scalability and decoupled energy-power design, yet they still face challenges in cost reduction and efficiency improvement, which necessitates advanced modeling to accelerate development as a complement to experiments. However, traditional numerical modeling is not efficient, restricting its application to optimal management. Artificial intelligence (AI) is revolutionizing this field by enabling accelerated simulations that integrate predictive accuracy and computational efficiency, while data-driven modeling empowers intelligent optimization of input design parameters. Beyond static modeling, AI techniques facilitate dynamic management through real-time state estimation and adaptive control strategies that respond to complex operating conditions. This review summarizes advances in recent five years of AI applications for flow batteries, and critically examine how the AI approaches address fundamental limitations in modeling and management paradigms, while identifying key challenges in model robustness and practical implementation that guide future research directions in developing intelligent flow battery systems.