Researchers in Japan say they have developed a method that improves wireless transfer systems, which could enable a fully wireless society. Credit: Volfgang (talk) – CC BY-SA 3.0 via Wkimedia Commons.
Researchers in Japan say they have developed a machine-learning method that significantly improves wireless power transfer systems, edging us closer to a “fully wireless society.”
The team from Chiba University reported this month in the journal IEEE Transactions on Circuits and Systems that their load-independent design keeps voltage stable even as devices draw varying amounts of current, a longtime technical hassle for wireless charging technology.
The study found the new approach reduced power fluctuations to about 5% compared with 18% for conventional, load-dependent systems, and raised transfer efficiency to as much as 86.7% from levels that can fall to 65%.
Wireless power transfer is already used in a variety of devices
Wireless power transfer, or WPT, is already used in devices ranging from smartphones and biomedical sensors to induction stovetops. But conventional systems often suffer when the electrical resistance of a receiving device changes during charging, causing voltage swings that can slow charging or damage batteries.
“Because the load changes as a battery charges, keeping voltage steady has been a major challenge,” said the study’s lead author, Hiroo Sekiya, a professor at Chiba University’s Graduate School of Advanced Integration Science. Sekiya said the machine-learning approach models the system virtually, runs simulations, and uses artificial intelligence to optimize performance by minimizing energy lost as heat and preserving signal quality.
WPT works through resonance, the same principle radio receivers use to pick out a station. A transmitter broadcasts an electromagnetic frequency tuned to a receiver, and if they resonate, energy transfer is amplified. The researchers’ technique trains an AI system to design resonance circuits that are less sensitive to what device or load is attached.
Japan’s breakthrough could also be important for complex batteries
The researchers said the improvement could be especially important for larger, more complex batteries such as those in electric vehicles, where charging loads can vary dramatically. Sekiya said the method also simplifies circuit construction and could reduce cost and size, with the team aiming to make wireless power commonplace within five to 10 years.
The study illustrates a broader trend in which artificial intelligence is applied to electrical-engineering problems, the authors said, noting the approach could automate aspects of circuit design and improve efficiency across power electronics.
While the Chiba team reported laboratory results showing higher efficiency and reduced fluctuations, broader commercial adoption will require further engineering, safety testing, and industry standards, experts say. The researchers suggested load-independent WPT could expand wireless power applications beyond consumer gadgets to electric vehicles and more complex systems.
AloJapan.com