In the 1990s, researchers undertook the Human Genome Project, a 13-year journey to enable a new era of innovation in medicine. By collecting and sharing vast amounts of data to uncover the secrets behind the genetic determinants of disease, this project has transformed the medical industry and led to countless breakthroughs. Now researchers are extending that innovation model to battery science with an eye on impact on a global scale.
Scientists from an international consortium led by researchers from the U.S. Department of Energy’s National Laboratories in Argonne and Idaho recently proposed a comprehensive new paradigm for data science called Battery data genome. It is an ambitious undertaking to develop unified data collection and sharing practices across the broad battery community. These innovative practices will create a comprehensive database network to enable breakthroughs in energy storage using artificial intelligence (AI).
“This is a call to action. We try to stimulate and organize the battery community to contribute their data, where possible to as many researchers as possible, to enable powerful data science methods to catalyze breakthroughs.” — Noah Paulson, battery scientist at Argonne
The electrochemical science that is urgently needed for a zero-carbon economy requires advanced data science,” said Sue Babinec, battery scientist at Argonne.“Addressing the highly complex technical questions facing battery scientists requires massive amounts of data to generate AI and machine learning algorithms.”
While there are a number of specialized battery data science projects, such as the Electrolyte Genome Project, the commitment to create a battery data genome dedicated to all aspects of the battery and unifying the work done by institutions and scales is unprecedented, said Argonne distinguished colleague and Joint Center for Energy Storage Research director George Crabtree.
According to Crabtree, the Battery Data Genome will collect and house data from every step of the battery lifecycle, from discovery to development to production and implementations of all kinds. Having universal data governance standards for every segment of the battery community is required for data creation to harness the power of AI algorithms designed to identify everything from new candidate electrode materials to improved battery construction to cell life.
This is a call to action,” says battery scientist Noah Paulson of Argonne.“We are trying to stimulate and organize the battery community to contribute their data, where possible to as many researchers as possible, to enable powerful data science methods to catalyze breakthroughs.”
According to Paulson, scientists are interested in many different characteristics and qualities when measuring the performance of a battery. As a result, the data sets collected by different groups, even those looking at the same battery in the same setup, will not be identical.“We need to find the basic set of information to associate with each set of data so that we don’t have to spend time cleaning the data to fit our models,” he said.
For batteries, there are many common types of data, but there is no uniform way to access them,” adds Argonne, computer scientist Logan Ward.“When data comes in many different formats, doesn’t mention how it’s collected, and isn’t often shared between different groups, it becomes very difficult to identify the kind of large-scale AI analysis and predictions needed to accelerate the development and deployment of new batteries.”
Having consistent and accessible data means formatting it in a specific way with unified metadata standards – specifying how the data is collected.“For example, metadata can include the ambient temperature or even the resistance of contacts to your electrodes,” said Babinec.
According to Paulson, better collaboration between the entire spectrum of battery researchers, from those who look at individual molecules to those who design and test battery packs, will be needed to create the new standards.
The transition of different groups of researchers studying different stages of battery development to create a universal set of data that can be widely accessed, understood and used is a major challenge, Babinec said.“It’s as if some of the community’s records were written in Spanish and some in German; you have to have a common language of science.”
To attract as many participants as possible, the Battery Data Genome offers many possibilities for sharing data.“It’s important to recognize that not all data needs to be shared openly for success; there are many different sharing scenarios that can provide individual benefits to the many groups in the complex ecosystem,” said Babinec.
This could potentially make participation in the Battery Data Genome more attractive to industrial partners, who could benefit from the data produced by academic or government partners without necessarily contributing their own data.“It’s no different from blood groups,” Babinec said.“Some groups would be universal data donors, other groups would be universal data recipients, and overall the community would benefit.”
Once scientists have filled the Battery Data Genome with data, they will have to test it. To do this, they will use“challenge problems” to validate the best AI algorithms using the data in the Battery Data Genome to solve real-world questions.“We may want to know what happens to a certain kind of battery that runs a certain number of cycles at a certain temperature, but to do that predictively using AI‘ said Babinec.“Having a strong repository of standardized data is the first step.”
“Having a standardized and easily accessible, comprehensive dataset can raise new questions for the battery community,” Crabtree said. “There are many unknown unknowns in batteries,” he said.“With access to data all meeting a universal set of standards, guided by machine learning and artificial intelligence, we can find new avenues for innovation that we haven’t considered until now,” he said.
Argonne already provides open software for cleaning existing data files with the “battery-data-toolkit” located at https://github.com/materials-data-facil ity/battery-datatolkit. A full file release for determining the life of 300 lithium-ion batteries with six different cathode chemistries will be released later in October.
A paper announcing the Battery Data Genome will appear in the October 19 issue of: Joules.
The Joint Research Center on Energy Storage (JCESR)a DO Energy Innovation Hub, is an important partnership that integrates researchers from many disciplines to overcome critical scientific and engineering barriers and create new breakthrough energy storage technology. led by the Argonne National Laboratory at the United States Department of Energy, partners include national science and engineering leaders from academia, the private sector and national labs. Their combined expertise spans the full range of the technology development pipeline, from basic research to prototype development to product engineering to market delivery.
Argonne National Laboratory seeks solutions to pressing national problems in science and technology. Argonne, the country’s first national laboratory, conducts industry-leading basic and applied scientific research in virtually every scientific discipline. Argonne researchers work closely with researchers from hundreds of companies, universities, and federal, state and municipal agencies to help them solve their specific problems, advance America’s scientific leadership, and prepare the nation for a brighter future. With employees from more than 60 countries, Argonne is managed by UChicago Argonne, LLC for the Office of Science of the United States Department of Energy.
Thanks to Argonne National Laboratory. By Jared Sagoff
Related story: Mapping the battery data genome for better batteries
Do you appreciate the originality of CleanTechnica and the coverage of cleantech? Consider becoming one CleanTechnica member, supporter, technician or ambassador — or a patron on Patreon.
Don’t want to miss a cleantech story? Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!
Do you have a tip for CleanTechnica, do you want to advertise or introduce a guest for our CleanTech Talk podcast? Contact us here.