UK researchers have developed a machine learning algorithm that could help reduce charging times and prolong battery life in electric vehicles by predicting how different driving patterns affect battery performance, improving safety and reliability.

The researchers, from the University of Cambridge, say their algorithm could help drivers, manufacturers, and businesses get the most out of the batteries that power electric vehicles by suggesting routes and driving patterns that minimize battery degradation and charging times.

The team developed a non-invasive way to probe batteries and get a holistic view of battery health. These results were then fed into a machine learning algorithm that can predict how different driving patterns will affect the future health of the battery.

If developed commercially, the algorithm could be used to recommend routes that get drivers from point to point in the shortest time without degrading the battery, for example, or recommend the fastest way to charge the battery without causing it to degrade.

The health of a battery, whether it’s in a smartphone or a car, is far more complex than a single number on a screen. “Battery health, like human health, is a multi-dimensional thing, and it can degrade in lots of different ways,” said first author Penelope Jones, from Cambridge’s Cavendish Laboratory. “Most methods of monitoring battery health assume that a battery is always used in the same way. But that’s not how we use batteries in real life. If I’m streaming a TV show on my phone, it’s going to run down the battery a whole lot faster than if I’m using it for messaging. It’s the same with electric cars – how you drive will affect how the battery degrades.”

“Most of us will replace our phones well before the battery degrades to the point that it’s unusable, but for cars, the batteries need to last for five, ten years or more,” said Dr. Alpha Lee, who led the research. “Battery capacity can change drastically over that time, so we wanted to come up with a better way of checking battery health.”

The researchers developed a non-invasive probe that sends high-dimensional electrical pulses into a battery and measures the response, providing a series of ‘biomarkers’ of battery health. This method is gentle on the battery and doesn’t cause it to degrade any further.

The electrical signals from the battery were converted into a description of the battery’s state, which was fed into a machine learning algorithm. The algorithm was able to predict how the battery would respond in the next charge-discharge cycle, depending on how quickly the battery was charged and how fast the car would be going the next time it was on the road. Tests with 88 commercial batteries showed that the algorithm did not require any information about the previous usage of the battery to make an accurate prediction.

The experiment focused on lithium cobalt oxide (LCO) cells, which are widely used in rechargeable batteries, but the method is generalizable across the different types of battery chemistries used in electric vehicles today.

“This method could unlock value in so many parts of the supply chain, whether you’re a manufacturer, an end user, or a recycler, because it allows capturing the health of the battery beyond a single number, and because it’s predictive,” said Lee. “It could reduce the time it takes to develop new types of batteries because we’ll be able to predict how they will degrade under different operating conditions.”

The researchers say that in addition to manufacturers and drivers, their method could be useful for businesses that operate large fleets of electric vehicles, such as logistics companies. “The framework we’ve developed could help companies optimize how they use their vehicles to improve the overall battery life of the fleet,” said Lee. “There’s so much potential with a framework like this.”

“It’s been such an exciting framework to build because it could solve so many of the challenges in the battery field today,” said Jones. “It’s a great time to be involved in the field of battery research, which is so important in helping address climate change by transitioning away from fossil fuels.”

The researchers are now working with battery manufacturers to accelerate the development of safer, longer-lasting next-generation batteries. They are also exploring how their framework could be used to develop optimal fast charging protocols to reduce electric vehicle charging times without causing degradation.

The research was supported by the Winton Programme for the Physics of Sustainability, the Ernest Oppenheimer Fund, the Alan Turing Institute, and the Royal Society.

Modelithics has introduced EMA Design Automation, Inc. (EMA) as a Modelithics Reseller.

“We are seeing dramatic growth in RF and microwave design content from our customers across the globe,” said Manny Marcano, President of EMA. “Partnering with Modelithics will allow us to provide our customers a plug-n-play solution to help them accelerate their RF analysis and ensure they are able to achieve first pass success for these critical RF systems and subsystems.”

As a reseller of Modelithics, EMA will be able to meet the needs of design engineers globally by offering high-accuracy RF and microwave active and passive simulation models for Modelithics' premium product the Modelithics COMPLETE Library, which includes models, representing more than 25,000 components from over 70 component and IC vendors. Also available are the mmWave & 5G Library, System Components LibraryT, and the COMPLETE+3D Library. The COMPLETE+3D Library includes a Modelithics extension collection of CLR component models, plus over 500 3D Geometry models. In addition, EMA will be representing Modelithics' broad array of the highest quality, RF/microwave/mm-wave Characterization and Modeling services, including Modelithics' world-class GaN modeling and 3D modeling capabilities.

“We welcome this new sales channel partnership with EMA as we bring together very complementary strengths for the benefit of our mutual customers,” said Larry Dunleavy, President, and CEO of Modelithics.

In planetary disks, carbon monoxide is lurking in large chunks of ice, solving the decade-old question, 'Where is the CO?' 

Astronomers frequently observe carbon monoxide in planetary nurseries. The compound is ultra-bright and extremely common in protoplanetary disks — regions of dust and gas where planets form around young stars — making it a prime target for scientists.

But for the last decade or so, something hasn't been adding up when it comes to carbon monoxide observations, says Diana Powell, a NASA Hubble Fellow at the Center for Astrophysics | Harvard & Smithsonian.

A huge chunk of carbon monoxide is missing in all observations of disks if astronomers' current predictions of its abundance are correct.

Now, a new model — validated by observations with ALMA — has solved the mystery: carbon monoxide has been hiding in ice formations within the disks. 

"This may be one of the biggest unsolved problems in planet-forming disks," says Powell, who led the study. "Depending on the system observed, carbon monoxide is three to 100 times less than it should be; it's off by a really huge amount."

And carbon monoxide inaccuracies could have huge implications for the field of astrochemistry.

"Carbon monoxide is essentially used to trace everything we know about disks — like mass, composition and temperature," Powell explains. "This could mean many of our results for disks have been biased and uncertain because we don't understand the compound well enough."

Intrigued by the mystery, Powell put on her detective hat and leaned on her expertise in the physics behind phase changes — when matter morphs from one state to another, like a gas changing into a solid.

On a hunch, Powell made alterations to an astrophysical model that's currently used to study clouds on exoplanets or planets beyond our solar system.

"What's really special about this model is that it has detailed physics for how ice forms on particles," she explains. "So how ice nucleates onto small particles and then how it condenses. The model carefully tracks where ice is, on what particle it's located on, how big the particles are, how small they are and then how they move around."

Powell applied the adapted model to planetary disks, hoping to generate an in-depth understanding of how carbon monoxide evolves over time in planetary nurseries. To test the model’s validity, Powell then compared its output to real ALMA observations of carbon monoxide in four well-studied disks — TW Hya, HD 163296, DM Tau and IM Lup.

The results and models worked really well, Powell says.

The new model lined up with each of the observations, showing that the four disks weren’t actually missing carbon monoxide at all — it had just morphed into ice, which is currently undetectable with a telescope.

Radio observatories like ALMA allow astronomers to view carbon monoxide in space in its gas phase, but ice is much harder to detect with current technology, especially large formations of ice, Powell says.

The model shows that, unlike previous thinking, carbon monoxide is forming on large particles of ice — especially after one million years. Prior to a million years, gaseous carbon monoxide is abundant and detectable in disks.

"This changes how we thought ice and gas were distributed in disks," Powell says. "It also shows that detailed modeling like this is important to understand the fundamentals of these environments."

Powell hopes her model can be further validated using observations with NASA’s Webb Telescope — which may be powerful enough to finally detect ice in disks, but that remains to be seen.

Powell, who loves phase changes and the complicated processes behind them, says she is in awe of their influence. "Small-scale ice formation physics influences disk formation and evolution — now that’s really cool."

Faculty in food science and computer science/computer engineering are collaborating to develop machine learning that can assist in the detection of autism spectrum disorder.

Could artificial intelligence be used to assist with the early detection of autism spectrum disorder? That’s a question researchers at the University of Arkansas are trying to answer. But they’re taking an unusual tack. 

Han-Seok Seo, an associate professor with a joint appointment in food science and the UA System Division of Agriculture, and Khoa Luu, an assistant professor in computer science and computer engineering, will identify sensory cues from various foods in both neurotypical children and those known to be on the spectrum. Machine learning technology will then be used to analyze biometric data and behavioral responses to those smells and tastes as a way of detecting indicators of autism.

There are several behaviors associated with ASD, including difficulties with communication, social interaction, or repetitive behaviors. People with ASD are also known to exhibit some abnormal eating behaviors, such as avoidance of some — if not many — foods, specific mealtime requirements, and non-social eating. Food avoidance is particularly concerning, because it can lead to poor nutrition, including vitamin and mineral deficiencies. With that in mind, the duo intends to identify sensory cues from food items that trigger atypical perceptions or behaviors during ingestion. For instance, odors like peppermint, lemons, and cloves evoke stronger reactions from those with ASD than those without, possibly triggering increased levels of anger, surprise, or disgust. 

Seo is an expert in the areas of sensory science, behavioral neuroscience, biometric data, and eating behavior. He is organizing and leading this project, including screening and identifying specific sensory cues that can differentiate autistic children from non-autistic children concerning perception and behavior. Luu is an artificial intelligence expert specializing in biometric signal processing, machine learning, deep learning, and computer vision. He will develop machine learning algorithms for detecting ASD in children based on unique patterns of perception and behavior in response to specific test samples. 

The duo is in the second year of a three-year, $150,000 grant from the Arkansas Biosciences Institute.

Their ultimate goal is to create an algorithm that exhibits equal or better performance in the early detection of autism in children when compared to traditional diagnostic methods, which require trained healthcare and psychological professionals doing evaluations, longer assessment durations, caregiver-submitted questionnaires, and additional medical costs. Ideally, they will be able to validate a lower-cost mechanism to assist with the diagnosis of autism. While their system would not likely be the final word in a diagnosis, it could provide parents with an initial screening tool, ideally eliminating children who are not candidates for ASD while ensuring the most likely candidates pursue a more comprehensive screening process.

Seo said that he became interested in the possibility of using multi-sensory processing to evaluate ASD when two things happened: he began working with a graduate student, Asmita Singh, who had a background in working with autistic students, and the birth of his daughter. Like many first-time parents, Seo paid close attention to his newborn baby, anxious that she be healthy. When he noticed she wouldn’t make eye contact, he did what most nervous parents do: turned to the internet for an explanation. He learned that avoidance of eye contact was a known characteristic of ASD. 

While his child did not end up having ASD, his curiosity was piqued, particularly about the role sensitivities to smell and taste play in ASD. Further conversations with Singh led him to believe fellow anxious parents might benefit from an early detection tool — perhaps inexpensively alleviating concerns at the outset. Later discussions with Luu led the pair to believe that if machine learning, developed by his graduate student Xuan-Bac Nguyen, could be used to identify normal reactions to food, it could be taught to recognize atypical responses, as well.

Seo is seeking volunteers 5-14 years old to participate in the study. Both neurotypical children and children already diagnosed with ASD are needed for the study. Participants receive a $150 eGift card for participating and are encouraged to contact Seo at hanseok@uark.edu.