Cost, performance prediction and optimization of a vanadium flow battery using machine learning. Credit: Li Tianyu.

Vanadium flow batteries (VFBs) are promising for stationary large-scale energy storage due to their high safety, long cycle life, and high efficiency.

The cost of a VFB system mainly depends on the VFB stack, electrolyte, and control system. Developing a VFB stack from lab to industrial scale can take years of experiments due to complex factors, from key materials to battery architecture.

Novel methods to accurately predict the performance and cost of a VFB stack and further system are needed in order to accelerate the commercialization of VFBs.

Recently, a research team led by Prof. Li Xianfeng from the Dalian Institute of Chemical Physics (DICP) of the Chinese Academy of Sciences proposed a machine learning-based strategy to predict and optimize the performance and cost of VFBs.

"We use AI technology to improve efficiency, reduce research time, and provide important guidance for the research and development of VFBs" said Prof. Li. "It may accelerate the commercialization of VFBs."

This work was published in Energy & Environmental Science on Sept. 22.

The proposed strategy takes operating as the main feature, and the material and structure of the stack as auxiliary features.

This can predict the voltage efficiency, energy efficiency, and electrolyte utilization ratio of the VFB stack, as well as the power and cost of the VFB system with high accuracy.

In addition, a future R&D direction for the VFB stack was proposed based on model coefficients of machine learning, i.e., developing high-power density VFB stacks under conditions of higher voltage efficiency and higher electrolyte utilization ratio.

This work not only has great significance for the R&D of VFB stacks, but also highlights the prospects for combining and experiments for optimizing and predicting the dynamic behavior of complex systems.

More information: Tianyu Li et al, Cost, Performance Prediction and Optimization on Vanadium Flow Battery by Machine-Learning, Energy & Environmental Science (2020). DOI: 10.1039/D0EE02543G

Journal information: Energy & Environmental Science