Visualize a scenario where a psychic tells the parents of a newborn child how long the child is going live. The same kind of experience is now possible for battery chemists who are making use of novel computational models to assess battery lifetimes depending on a single cycle of experimental data.
Argonne researchers have used machine learning models to make predictions of battery cycle life for a wide range of different chemistries. Image Credit: Shutterstock/Sealstep.
Investigators from the US Department of Energy’s (DOE) Argonne National Laboratory have resorted to the power of machine learning to forecast the lifetimes of an extensive range of various battery chemistries.
It is possible for researchers to precisely identify just how long various batteries will continue to cycle with the help of experimental data collected at Argonne from a set of 300 batteries which represented six different battery chemistries.
As far as a machine learning algorithm is concerned, researchers train a computer program to create inferences on an initial set of data and subsequently take what it has absorbed from that training to make decisions on one more set of data.
For every different kind of battery application, from cell phones to electric vehicles to grid storage, battery lifetime is of fundamental importance for every consumer. Having to cycle battery thousands of times until it fails can take years; our method creates a kind of computational test kitchen where we can quickly establish how different batteries are going to perform.
Noah Paulson, Study Author and Computational Scientist, Argonne National Laboratory
“Right now, the only way to evaluate how the capacity in a battery fades is to actually cycle the battery. It’s very expensive and it takes a long time,” added Argonne electrochemist Susan “Sue” Babinec, the study’s co-author.
Paulson feels that the process of fixing a battery life can be a difficult task.
“The reality is that batteries don’t last forever, and how long they last depends on the way that we use them, as well as their design and their chemistry. Until now, there’s really not been a great way to know how long a battery is going to last. People are going to want to know how long they have until they have to spend money on a new battery,” Paulson added.
One special aspect of the study is that it depended on comprehensive experimental work performed at Argonne on a range of battery cathode materials, particularly Argonne’s patented nickel-manganese-cobalt (NMC)-based cathode.
We had batteries that represented different chemistries, that have different ways that they would degrade and fail. The value of this study is that it gave us signals that are characteristic of how different batteries perform.
Noah Paulson, Study Author and Computational Scientist, Argonne National Laboratory
According to Paulson, additional studies in this area could guide lithium-ion batteries’ future.
One of the things we’re able to do is to train the algorithm on a known chemistry and have it make predictions on an unknown chemistry. Essentially, the algorithm may help point us in the direction of new and improved chemistries that offer longer lifetimes.
Noah Paulson, Study Author and Computational Scientist, Argonne National Laboratory
In this approach, Paulson is convinced that the machine learning algorithm could expedite the development and testing of battery materials.
Paulson says, “Say you have a new material, and you cycle it a few times. You could use our algorithm to predict its longevity, and then make decisions as to whether you want to continue to cycle it experimentally or not.”
Babinec adds, “If you’re a researcher in a lab, you can discover and test many more materials in a shorter time because you have a faster way to evaluate them.”
The study was published in the Journal of Power Sources.
Besides Paulson and Babinec, the other authors of the study include Joseph Kubal, Logan Ward, Saurabh Saxena, and Wenquan Lu from Argonne.
This study was financially supported by an Argonne Laboratory-Directed Research and Development (LDRD) grant.
Journal Reference:
Paulson, N. H., et al. (2022) Feature engineering for machine learning enabled early prediction of battery lifetime. Journal of Power Sources. doi.org/10.1016/j.jpowsour.2022.231127.