Page 50 - FDMAsia NovDec 2024

48 MATERIALS NOV/DEC 2024 FDM ASIA | www.fdmasia.com

Liji Jinaraj ascertain the contact properties.
In terms of the wood species used, spruce was the best
solution considering that it is a common species used for the
realisation of glued laminated timber elements according to the

study of "Metamodel assisted optimisation of glued laminated
timber beams by using metaheuristic algorithms" from Pech,
S.,Kandler, G., Lukacevic, M. and Füss, J..
The GLT elements are formed by joining together layers
of wood material in the form of lamellas with the thickness
=2 cm with a structural epoxy resin-based adhesive.
The adhesive used for processing the glued laminated
timber elements, Melamine-urea-formaldehyde (MUF), is one
resources, whether in laboratory experiments or through time- of the most commonly used adhesives in the wood industry.

consuming programs. A representation of a glued laminated timber element can be
This happens considering that the operation research seen in “representation of a physical model of a glued laminated
processing on models using the Finite Element Method (FEM) timber element with a section representing the position of the
involves systems solving inequations having as input data fibres in the element” where the lamellas can be identified.
the results of the FEM and the geometry variation until the However, the adhesive is not visible in the glued laminated
objective function is reached. timber section being a part of the element. The element is
This is the current scientific context, where the premises good quality which means that poor glue bonding between

exist for the use of machine learning techniques in order the layers is not encountered so the section is evenly created.
to embed geometry and structural behaviour data obtained Given the natural composition of wooden material in the
by topological optimisation and to reduce the computation form of annual growth rings, the position of the fibres is
requirements for the deployment of such a solution. essential when using the material in a structure. The mechanical
Techniques such as deep neural networks or generative properties are different along the axis of the element and
adversarial networks have already proven useful for this kind for maximum utilisation of it they have to be parallel to the
of application. The part this paper addresses is the training longitudinal axis.
data for such an algorithm. The laboratory determinations were carried out in two
The success of any development in this direction depends stages: in the first one, the strength parameters of glued

on the quality of the training data; for example, the present laminated timber were directly determined, and in the second
project proposes some domain limits for the exploration of one, the shear strength parameters on the glulam material
topologically optimised solutions and presents an algorithm for were determined.
computing this kind of optimisation task, using the Dassault The tests revealed values much higher than the shear
ABAQUS 2023 software package. strength of wood material, which makes it so that the FEM
modelling of the glulam-type element can be achieved either in
Experimental Campaign And Test Equipment the form of homogeneous material or using “tie”-type contact

For the experimental campaign, GLT elements were used in elements. The experimental campaign was conducted in the
the experiments and tested as well as their combination to laboratories of the Technical University of Civil Engineering