Machine Learning can predict the lifetime of a battery

Researchers at the U.S. Department of Energy’s Argonne National Laboratory have used machine learning to predict the lifetimes of a wide range of different battery chemistries. By using experimental data gathered from a set of 300 batteries the scientists can accurately determine just how long different batteries will continue to cycle. Battery lifetime is of fundamental importance for every consumer. Currently the only way to work out how the capacity in a battery, is to actually cycle the battery. 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. the studies have shown that machine learning tools can predict lithium-ion battery lifetimes from 100 or fewer preliminary cycles, representing only a few weeks of cycling and with this information new and better designs can be made.

Noah H. Paulson et al.; Feature engineering for machine learning enabled early prediction of battery lifetime; Journal of Power Sources; Volume 527, 15 April 2022, 231127

Photo Pixabay


Comments

Post a Comment

Popular posts from this blog

New photocatalyst made from an aerogel for more efficient hydrogen production

Image
  Researchers at ETH Zurich have developed a new photocatalyst made from an aerogel that could enable more efficient hydrogen production. One use for aerogels based on nanoparticles is as photocatalysts. These are employed whenever a chemical reaction needs to be enabled or accelerated with the aid of sunlight – one example being the production of hydrogen. The material of choice for photocatalysts is titanium dioxide (TiO 2 ), a semiconductor. But TiO 2  has a major disadvantage: it can absorb only the UV portion of sunlight – just about 5 percent of the spectrum. If photocatalysis is to be efficient and industrially useful, the catalyst must be able to utilise a broader range of wavelengths. Junggou Kwon has been looking for a new way to optimise an aerogel made of TiO 2  nanoparticles. And she had a brilliant idea: if the TiO 2  nanoparticle aerogel is “doped” (to use the technical term) with nitrogen, such that individual oxygen atoms in the material are repla...
Read more

Spider webs near roads capture all types of Micro plastics and could be used for monitoring pollution

Image
Researchers at the University of Oldenburg have been examining spider webs webs for the smallest plastic particles – on inner-city streets with varying levels of traffic. They found mainly the plastic PET, presumably from textiles, as well as particles from the abrasion of car tyres and polyvinyl chloride (PVC). The amounts of plastic particles found depended on the location. Spider webs, the team concludes, are a simple and inexpensive means of monitoring air pollution by microplastics in the city and identifying particularly polluted areas.  The samples, collected in a mid-sized German city, were processed with  Fentons  reagent and measured using pyrolysis-gas chromatography–mass spectrometry for specific, polymer related indicator compounds. All samples contained    microplastics (MP) including tyre wear particles (TWP) contamination in air samples.  All the spider webs were contaminated with microplastics. In some cases, the plastic content even accoun...
Read more

​ New artificial leaf can capture 100 times more carbon in normal conditions

Image
  Engineers at the University of Illinois have managed to construct a cost-effective artificial leaf that can capture carbon dioxide 100 times better than any current system works in normal air which is containing very dilute sources of carbon dioxide. This artificial leaf is a modified version of existing systems with a built-in water gradient creating a dry on a wet side across a charged membrane. On the dry side and organic solvent attaches to the carbon dioxide producing a concentrated by carbonate solution. As the bicarbonate solution increases these are pulled across the membrane towards a positively charged electrode in a water-based solution. The liquid solution dissolves the HCO3 back into carbon dioxide so that it can be released and harnessed for fuel uses. The current system is small enough to fit in a backpack and uses less than the electricity used by 1 W LED lightbulb. The system is stackable so that modules can be added or subtracted and could be affordable are used...
Read more