press release
Published: 16 November 2026

Surrey framework turns everyday vibrations into power and information

New technology could allow wearables to one day rely less on batteries and more on vibrations generated by the user. The model, developed by the University of Surrey and its spin-out company Z-PULSE, predicts how much electricity small generators can make from vibrations. 

Researchers used triboelectric nanogenerators (TENGs), a type of device that makes electricity when two materials come into contact and separate – similar to when the static energy you get when rubbing a ballon on a jumper. Contact leaves the surfaces charged, and moving them apart pushes charge around a circuit. TENGs are light and flexible and can work with slow, irregular motion. Predicting what they will deliver has been harder, because many circuit models treat them as purely electrical parts and overlook how they move. 

The Surrey team has developed a lumped electromechanical equivalent-circuit framework, which represents the device’s mechanical and electrical behaviour in one model. It accounts for factors including mass, stiffness, energy losses, electrostatic forces, the changing gap between charged surfaces and the connected electrical load. 

The model enables engineers to analyse the complete path from vibration to electrical output and identify where energy is gained, transferred or lost. It also makes it possible to design around the device’s mechanical resonance, the frequency at which it naturally vibrates most strongly, rather than optimising materials or electrical components in isolation.  

The results put tuning to the right vibration at the centre of good design. In simulations, changing the mass of one design by about 3 per cent moved it off resonance and cut its peak power roughly 60-fold. That is because matching a device's natural frequency to the vibration around it builds motion and electrical output, much as pushing a swing in time builds height.

The researchers also investigated whether the electrical signal could provide information about the movement that produced it. They mounted a TENG on a bicycle and recorded its response on smooth, rough and gravel surfaces. Each surface generated a distinct electrical pattern. A machine-learning model identified the surface with 96.7 per cent accuracy. This suggests that a single device could both harvest energy and act as a sensor, for example, in future autonomous monitoring applications.

The study has been published in Advanced Energy Materials. 

###

Notes to editors 

Media Contacts


External Communications and PR team
Phone: +44 (0)1483 684380 / 688914 / 684378
Email: mediarelations@surrey.ac.uk
Out of hours: +44 (0)7773 479911