18 December 2024
New research shows how AI can take construction materials to a new level
A new doctoral thesis from Luleå University of Technology shows how data-driven analysis can improve the use of cross-laminated timber (CLT) in construction. The research challenges traditional approaches and opens new ways of achieving better material performance and a more efficient use of wood resources, including wood previously considered unsuitable for sustainable structural applications.
Dietrich Buck’s thesis integrates established and innovative methods to improve the functionality of CLT. By combining digital image correlation, hyperspectral imaging, and AI, the work provides a deeper insight into the mechanical properties of materials.
"We have shown that knots in specific layers of CLT, previously considered a weakness in wood structures, can actually improve shear resistance," says Dietrich Buck, doctoral student in Wood Science and Engineering at Luleå University of Technology.
The thesis also explores how techniques such as four-dimensional X-ray computed tomography and NIR spectroscopy can be used to monitor and improve wood’s resistance to moisture and mould fungus. By integrating these techniques with AI, it has become possible to detect hidden patterns and properties that traditional methods fail to capture.
Dietrich Buck, PhD student in Wood Science and Engineering at Luleå University of Technology.
The thesis also reveals the potential to change industry standards for timber. Wood previously considered unsuitable for structural purposes may now be utilised, reducing waste, thus increasing the availability of construction materials. "We have developed an assessment framework which is application-independent and integrates measurement technologies with AI tools to create a more complete understanding of composite materials such as CLT," Buck explains.
This research has the potential for practical applications, with the possibility of serving as a quality control system in the industry. By combining data from different measurement technologies, the properties of materials can be optimised during the processing phase. These techniques also have the potential to improve the cutting and edge-sealing of CLT to reduce the risk of mould fungus.
"This research demonstrates how data-driven methods and data fusion can raise materials science to a higher level. By combining traditional principles with modern technology, we are opening up new ways to create sustainable building technologies," Buck concludes.
Contact
Dietrich Buck
- Doctoral student
- 0910-585729
- dietrich.buck@associated.ltu.se
- Dietrich Buck
Published:
Updated: