Machine learning for flow batteries: opportunities and challenges

Li, Tianyu; Zhang, Shun; Li, Xianfeng

doi:10.1039/d2sc00291d

Cited by 23 publications

(18 citation statements)

References 111 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…63,66,68,72,75,77 Accurate measurements of RAOM stability are critical, not only to lifetime cost analysis of potential candidates for decadal RFB operation, 10 but also to the community-wide development of rich data sets of RAOM lifetimes. Such knowledge can complement high-throughput theoretical screening, [78][79][80][81][82][83] enhance machine learning capabilities, 84,85 and motivate RAOM stability prediction -which still remains an open problem.…”

Section: Resultsmentioning

confidence: 99%

High-Throughput Electrochemical Characterization of Aqueous Organic Redox Flow Battery Active Material

Fell,

Aziz

2023

J. Electrochem. Soc.

View full text Add to dashboard Cite

The development of redox-active organics for flow batteries providing long-discharge-duration energy storage requires an accurate understanding of molecular lifetimes. Herein we report the development of a high-throughput setup for the cycling of redox flow batteries. Using common negolyte redox-active aqueous organics, we benchmark capacity fade rates and compare variations in measured cycling behavior of nominally identical volumetrically unbalanced compositionally symmetric cells. We propose figures of merit for consideration when cycling sets of identical cells, and compare three common electrochemical cycling protocols typically used in battery cycling: constant current, constant current followed by constant voltage, and constant voltage. Redox-active organics exhibiting either high or low capacity fade rates are employed in the cell cycling protocol comparison, with results analyzed from over 50 flow cells.

show abstract

Section: Resultsmentioning

confidence: 99%

High-Throughput Electrochemical Characterization of Aqueous Organic Redox Flow Battery Active Material

Fell,

Aziz

2023

J. Electrochem. Soc.

View full text Add to dashboard Cite

show abstract

“…63,66,68,70,72 Accurate measurements of RAOM stability are critical, not only to lifetime cost analysis of potential candidates for decadal RFB operation, 10 but also to the community-wide development of rich data sets of RAOM lifetimes. Such knowledge can complement high-throughput theoretical screening, [73][74][75][76][77][78] enhance machine learning capabilities, 79,80 and motivate RAOM stability prediction which still remains an open problem.…”

Section: Cycling Protocol Comparisonmentioning

confidence: 99%

High-Throughput Electrochemical Characterization of Aqueous Organic Redox Flow Battery Active Material

Fell

Aziz

2023

Preprint

View full text Add to dashboard Cite

The development of redox-active organics for flow batteries providing long duration energy storage requires an accurate understanding of molecular lifetimes. Herein we report the development of a high-throughput setup for the cycling of redox flow batteries. Using common negolyte redox-active aqueous organics, we benchmark capacity fade rates and compare variations in measured cycling behavior of identical volumetrically unbalanced compositionally symmetric cells. We propose figures of merit for consideration when cycling sets of identical cells, and compare three common electrochemical cycling protocols typically used in battery cycling: constant current, constant current constant voltage, and constant voltage. Redox-active organics exhibiting either high or low capacity fade rates are employed in the cell cycling protocol comparison, with results shown from over 50 flow cells.

show abstract

“…The models were validated by checking the performance accuracy on the train and test sets. The validated model was then used to predict data properties [20]. For model evaluation, the DFT calculated properties were compared with the corresponding machine learning predicted values.…”

Section: Model Validation and Evaluationmentioning

confidence: 99%

Machine learning models for predicting density of sodium-ion battery materials

Monareng,

Maphanga,

Ntoahae

2023

MATEC Web Conf.

View full text Add to dashboard Cite

With the unprecedented amounts of material data generated from high-throughput density functional theory, machine learning provides the ability to accelerate the discovery and design of new materials. In this work, machine learning regression techniques are applied to a large amount of data from Materials Project Database, to develop machine learning models capable of accurately predicting the densities of sodium-ion battery cathode materials. Different machine learning regression models are successfully developed and validated. Feature vectors derived from the properties of materials’ chemical compounds are evaluated. Extra trees regressor model is found to be the best model in predicting the density with an accuracy of 0.95 and 0.09 g/cm3 coefficient of determination and mean square error, respectively.

show abstract

Machine learning for flow batteries: opportunities and challenges

Abstract: With increased computational ability of modern computers, the rapid development of mathematical algorithms and continuous establishment of material databases, artificial intelligence (AI) has shown tremendous potential in the chemistry field....

Cited by 23 publications

References 111 publications

High-Throughput Electrochemical Characterization of Aqueous Organic Redox Flow Battery Active Material

High-Throughput Electrochemical Characterization of Aqueous Organic Redox Flow Battery Active Material

High-Throughput Electrochemical Characterization of Aqueous Organic Redox Flow Battery Active Material

Machine learning models for predicting density of sodium-ion battery materials

Contact Info

Product

Resources

About