Better Touch Better Business
Contact Sales at UFO.
The market for artificial intelligence (AI) technology is drawing people’s attention and the rapidly developing technology of artificial intelligence and machine learning are transforming business and scientific research. With recent research, researchers has taken advantage of the latest AI technology for lithium-ion battery.
A team at Imperial College London has developed a new machine learning algorithm allowing them to explore the microstructure of fuel cells and lithium-ion batteries, and run 3D simulations to predict and improve cell performance on the basis of their microstructure.
The microstructure of lithium-ion batteries and fuel cells both closely impacts their performance: the shape and arrangement of the pores(holes) inside their electrodes can have an impact on the amount of power that fuel cells can generate, and the charge and discharge time of lithium-ion batteries.
Due to the very small micrometer-scale pores of fuel cells and lithium-ion batteries, it is difficult to examine their specific shapes and size at a resolution that is high enough to tie them to overall performance of the cell.
The innovative machine learning technique that the researchers used is known as “ deep convolutional generative adversarial networks” (DC-GANs). These algorithms are able to learn to generate 3D image data of the microstructure of cells based on the training data gained from nano-scale imaging performed synchrotrons, which is a type of particle accelerator that is similar to the size of a football stadium. The research has been recently published in npj Computational Materials.
By using this new machine learning method, smartphones could be charged faster, the time for using electric vehicles could last longer, and the power of hydrogen fuel cells running data centers is boosted.
“Our technique is helping us zoom right in on batteries and cells to see which properties affect overall performance. Developing image-based machine learning techniques like this could unlock new ways of analyzing images at this scale.”
-- From Andrea Gayon-Lombardo, Lead author, Imperial’s Department of Earth Science and Engineering
“Our team’s findings will help researchers from the energy community to design and manufacture optimized electrodes for improved cell performance. It’s an exciting time for both the energy storage and machine learning communities, so we’re delighted to be exploring the interface of these two disciplines.”
-- From Dr. Sam Cooper, Project supervisor, Imperial’s Dyson School of Design Engineering
With the improvement in battery research, this innovative machine learning algorithm is likely to be applied into the manufacturing of batteries, making the design of optimized electrodes for next-generation cells become possible.
Gayon-Lombardo, A., et al. (2020) Pores for thought: generative adversarial networks for stochastic reconstruction of 3D multi-phase electrode microstructures with periodic boundaries. npj Computational Materials. doi.org/10.1038/s41524-020-0340-7.
Hello, please leave your name and email here before chat online so that we won't miss your message and contact you smoothly.