Finger-joint optimization of spruce lumber using finite element and statistical modelling

Said, Mohamed Essam Abdelazim Ahmed (2018) Finger-joint optimization of spruce lumber using finite element and statistical modelling. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (6MB)

Abstract

The primary objective of this thesis was to investigate and explore the feasibility of using a numerical 3-D finite element analysis (FEA) and statistical design of experiment methodology (DOE) to optimize finger-joint (FJ) configuration. This thesis provides guidelines to achieve structural behaviour of finger-joined elements that approaches non-finger-joined (NFJ) lumber. The optimum configurations will improve strength, save time, and money. A numerical 3-D finite element model was established using Abaqus FEA to simulate the interface between the two FJ adherent parts. The model was verified with experimental tests of actual FJ samples which showed a close agreement with the corresponding numerical model. An optimization of FJ lumber in horizontal, vertical and slope orientations were carried out to obtain the FJ geometry that achieves maximum structural performance of such lumber. A Reduced Quadratic Response Surface Model (RSM) and a Modified Two-Factor Interaction (2FI) experimental design were then used for both Normal and Inclined FJ categories, respectively. The statistical model were set up to assess the effect of the main factors: FJ-length, FJ-pitch, and FJ-tip width (tip thickness) and their interactions values on the stiffnesses. One additional parameter, FJ-Slope of three-levels was added with the Inclined FJ category. The obtained results showed clearly the potential of increasing FJ resistance by optimizing its geometry. In addition, it was observed that FJ in vertical orientation achieved higher structural performance close to the NFJ lumber comparing to the FJ in the horizontal orientation. Regression analysis was used to develop a design equations per each joint orientation (horizontal and vertical). The predicted equations will be useful to determine the optimum FJ geometry directly without the need to use a trial and error approach to achieve structural behaviours close to the NFJ lumber. Finally, enhancing these properties impact the behaviour and capacity of finger-joined elements at the serviceability and ultimate limit states.

Actions (login required)

View Item