The application of finite element techniques in the design of timber structures

Abstract

Structural timber exhibits significantly greater variability in strength properties, both within individual members and between different members, particularly when compared to other structural materials. Eurocode 5, the code which governs the design of timber structures, is based on the limit state concept combined with a partial factor method. From previous studies and tests which have been carried out in the past, discrepancies arise from the predictions to the outcome results of tests done in the laboratory. This may be the cause of several factors which the code does not consider such as defects in timber, openings made within the element and its drying process which significantly impact the material's final strength. Through the use of Finite Element software-based design one can model the ideal timber structure, analyse it and design it. In fact, after conducting a comprehensive analysis and comparison of the results, a statistical model can be developed to predict the structural behavior of timber elements. This dissertation investigates the accuracy of reproducing laboratory test results for compression, tension and bending loads performed on timber structures by Farrugia (2023). Average results from laboratory tests were modelled in the FE program LUSAS (2020). The curves obtained were plotted against the averaged laboratory results to confirm accuracy. The final results obtained for the compression, tension and bending tests were then compared to the University of Stuttgart, 'Guidelines for a Finite Element-Based Design of Timber Structures' (2022). The FEA results aligned well with the laboratory results in all tests, resulting in a relative difference of less than 0.5 in all tests performed, suggesting that the models comply with the guidelines in terms of overall behaviour.