As mentioned at the outset, wood has complex mechanical behaviour with highly nonlinear performance [
11]. However, in existing studies, wood is sometimes simulated as a homogeneous and isotropic material. This assumption will not be able to illustrate the practical mechanics performance [
12]. For analysing the failure and fatigue behaviour of wood structures like piles or poles, it is well recognised that it is necessary to model wood as an orthotropic material with elastic-plastic behaviour based on Hill’s criterion [
13,
14]. Besides, Piao developed FE models to forecast the performance of the uniform-diameter wood poles subjected to the loads [
15]. Five orthotropic models were developed using ANSYS and were verified with experimental results. In the research, three-dimensional (3D) 10-node tetrahedral solid element was used in the modelling of wood laminated poles, which has a nonlinear displacement behaviour, plasticity, large deflection, and large strain capabilities. The predicted deflection by these models agreed well with those of the experiments, and the predicted normal stress agreed with those calculated. Pellicane and Franco employed 3D FE models to investigate the stress distribution and failure of wood poles subjected to cantilever bending [
16]. Orthotropic and isotropic material models with linear-elastic material properties were built and verified with the theoretical results. It can be concluded that orthotropic models are more sensitive to support conditions affecting the bending stress distribution near the boundary. Bulleit and Falk developed an isotropic material model to investigate the usage of guided wave velocity measurements to distinguish between strength-reducing decay and non-strength-reducing growth ring separations (ringshake). The results from the field testing and FE analysis indicated that the guided wave velocity measurements alone were not sufficient to confidently differentiate between ringshake and decay [
17].
On the other hand, the soil is also a complex material, which consists of particles with various mineralogy, size and shape [
18]. Therefore, soil modelling plays an essential role in pile or pole testing research, because the piles and poles are generally embedded in the soil, the boundary condition of which will directly affect the wave propagation behaviour in structures when stress wave-based NDE techniques are adopted for tests. High strain testing of piles or poles is typically performed to evaluate the driving system for assessing the static or bearing capacity. For this kind of testing, soil behaviour is always considered as elasto-plastic and simulated by the Mohr-Coulomb model to describe the stability of foundation structures. Abdel-Rahman and Achmus created numerical models to illustrate the behaviour of medium dense sand when the piles are under an inclined load [
19]. Choi et al. used a Mohr-Coulomb elasto-plastic model to describe the soil behaviour and pile foundation stability when a strong earthquake occurred [
20]. Moreover, low strain methods have also been developed for the integrity assessment of embedded length estimation of piles. For a very low strain level (less than 10
−5), the elastic and linear behaviour of soil is obvious [
21]. Accordingly, soil could be simulated as linear-elastic, which provides the most basic soil behaviour without considering the nonlinear and plastic behaviour at failure of the soil. Chow et al. adopted a linear-elastic model to describe the soil behaviour under integrity testing [
22]. In this work, low strain testing is conducted and the strain level is less than 10
−5. Consequently, the material constitutive in this paper will be simulated as linear-elastic. Furthermore, to effectively analyse the wave propagation in wood poles, the effect of soil-pole interaction should be also considered. For the current research on soil-pile/pole interaction, interfaces are commonly simulated in two ways, perfectly bonded interface or frictional interface. The frictional interface allows slipping and gapping between the soil and the structure, more accurately representing the practical behaviour. In this case, the Coulomb’s friction model is employed to simulate slipping and gapping in FE analysis, but it will take more calculation time and computer resources. Accordingly, if slipping and gapping is negligible or will not affect the analysis result, perfect bonding using the coupling method can be applied [
23]. In the field of pile/pole testing, the soil is also considered as the linear-elastic material, the plastic deformation of which can be depicted by a series of linear springs. In this case, the slip effect between structure and soil can be neglected [
24]. In ANSYS, interaction behaviour modelling can be achieved via contact analysis considering different contact conditions. Ji and Wang adopted surface contact element in ANSYS to simulate the integrity testing of a wharf pile based on one-dimensional guided wave theory [
25]. The results from numerical results matched with experimental results quite well and the defects of the wharf piles could be detected accurately. The dynamic response of wharf piles caused by pulse loading can be accurately simulated by ANSYS. Interface elements (contact pairs) in ABAQUS were chosen by Miao et al. to investigate the response of a single pole when subjected to lateral soil movements [
26]. These elements were allowed to separate if there was tension across the interface and the shear and normal force are set to zero once a gap is formed. Ninić et al. studied the pole-soil interaction along the pole skin under an axial loading using a 3D frictional point-to-point contact formulation, where each pole integration point was associated with an adjacent target point in the soil element [
27]. It has been shown that pole responses are comparable with the analytical models he developed.