Engineering news
The new ‘optimisation framework’ developed by the team could also reduce the cost of assessments for new designs, they said.
"The goal is to design a better battery and, traditionally, the industry has tried to do that using trial and error testing," said mechanical engineer and research leader Wei Lu. "It takes such a long time to evaluate."
Parameters involved in battery design include the materials used, thickness of the electrodes, size of particles in the electrodes and more. Testing each configuration usually means several months of fully charging then fully discharging – cycling – the battery, mimicking a decade of use.
The UM engineers harnessed machine learning to create a system that ‘knows’ when to quit and how to improve over time. The framework halts cycling tests that do not have a promising start in order to save resources. It also takes data from previous tests and suggests new sets of promising parameters to investigate.
"Our approach not only reduces testing time, but it automatically generates better designs," Lu said. "We use early feedback to discard unpromising battery configurations rather than cycling them 'til the end. This is not a simple task, since a battery configuration performing mediocrely during early cycles may do well later on, or vice versa. We have... enabled the system to learn from the accumulated data to yield new promising configurations."
As well as stopping tests on designs that lack promise, the system also generates multiple configurations to be tested simultaneously, known as asynchronous parallelisation. If any configuration completes testing or is discarded, the algorithm immediately calculates a new configuration to test without needing to wait for the results of other tests.
"This framework can be tuned to be more efficient when a performance prediction model is incorporated," said Changyu Deng, first author of a paper on the work. "We expect this work to inspire improved methods that lead us to optimal batteries to make better EVs and other life-improving devices.”
The speed of testing “could provide a major boost to battery developers searching for the right combination of materials and configurations to ensure that consumers always have enough capacity to reach their destinations”, a research announcement said.
"By significantly reducing the testing time, we hope our system can help speed up the development of better batteries, accelerate the adoption or certification of batteries for various applications, and expedite the quantification of model parameters for battery management systems," Lu said.
The research was funded by LG Energy Solution. It was published in Patterns-Cell Press.
Get to grips with the future factory at Advanced Manufacturing (18-22 July), part of the Engineering Futures webinar series. Register for FREE today.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.