A recent [Nature] publication from researchers at Stanford University, MIT, and Toyota describes how machine learning models can be used to predict the useful life of advanced lithium-ion and any future "next-generation" battery chemistry. [News Article from Stanford] The research team used Arbin battery test equipment over the past two years for the on-going study.
About the Battery Research Project
Batteries were tested to end-of-life and then machine learning was used to analyze the Arbin data and develop algorithms to predict battery life based on early cycles. Charge-discharge cycling to end-of-life can take months or years for advanced chemistry batteries comprising many thousands of cycles. It is a slow and costly phase of the battery research and development process. Expediting this process by identifying key metrics and indicators in the data during early cycles (<100) is critical to reduce the time required for battery development. Beyond material development and cell-grading, these evaluation testing techniques can also be used to evaluate fast-charge protocols, which is the next phase of the research project.
The joint team has published data that is available publicly [https://data.matr.io/1/].
Arbin [LBT test equipment] and [cell holders] were used along with temperature chambers.
Request a Quote or Contact us to learn more: email@example.com | +1 979 690 2751 | www.arbin.com
Why Arbin is Most Suitable for This Level of Battery Research
Previous research has been done using coulombic efficiency as the primary metric to predict battery end-of-life [Source]. Arbin was involved in a 3-year [ARPA-E project] from 2012 through 2015 with Ford Motors and Sandia National Lab to develop a new generation of high-precision battery test systems that are capable of performing meaningful coulombic efficiency calculations on high-current cells. This technology has been implemented across [Arbin’s cyclers] and is available to researchers worldwide.
The new findings from the team at Stanford, MIT, and Toyota is another breakthrough that has utilized Arbin’s 24-bit resolution and superior measurement precision. Arbin has also recently developed a new [cell-isolating thermal safety chamber], "MZTC," that has 8 independently controlled mini-chambers in 1. It allows greater temperature uniformity by isolating individual or small sets of cells that can reduce the error calculation and further improve the machine learning algorithms to evaluate battery life. It also provides a safer environment when testing cells at high c-rates.
Arbin is committed to providing the best test equipment as a tool for researchers because we understand the import role energy storage plays in our everyday life and future. Battery test equipment is available for [materials research applications], up to [commercial cell testing at high c-rates]. [Arbin’s cell-isolating thermal safety chamber] also provides a greater temperature stability and uniformity than a traditional large chamber can provide.