Researchers have introduced a new approach to visualizing acoustic emission (AE) that offers deeper insight into how damage develops in bio-based composite materials. The findings point toward more reliable structural health monitoring and improved prediction of component behavior under real service conditions.
Basalt bio-epoxy laminates are gaining traction in sustainable construction and industry because they combine strong mechanical performance with a reduced environmental footprint. Their durability, however, still depends on complex internal damage processes that evolve as the material is loaded.
Conventional monitoring techniques are often not sensitive enough, or they require invasive inspection. To address this, the research team tested laminated specimens with different fiber orientations using a purpose-built Arcan fixture, while simultaneously recording acoustic emission (AE) signals.
A key contribution of the work is an ellipsoidal representation of AE signals, which provides a more intuitive view of signal magnitude and scatter. This makes interpretation easier and improves the reliability of early damage detection.
Using artificial intelligence—specifically neural networks—the researchers also developed classifiers that can successfully categorize AE signals according to load level. This supports better predictions of when and how the material is likely to fail.
“Our approach brings together precise mechanical testing, advanced signal processing, and machine learning. The result is a tool that can help engineers monitor structural integrity in real time,” explains Assist. Prof. Tomaž Kek, one of the study’s authors.
The methodology strengthens the development of predictive models for composite behavior under load and represents an important step toward safer and more sustainable use of these materials in practice.
Full paper: Ellipsoidal acoustic emission patterns in basalt bio-epoxy laminates under different loading angles— Tomaž Kek, Zoran Bergant, Roman Šturm (IF: 6.0).