Zhenmin Zou

Modelling Interlaminar and Intralaminar Damage in Filament-Wound Pipes under Quasi-Static Indentation

A pragmatic modelis proposed to predict interlaminar and intralaminar damage, i.e., delamination and matrix cracking, in filament wound pipes. The pipe is modelled as an assembly of sublaminates connected along their interfaces. The initiation of delamination and transverse matrix cracking is predicted based on stress-based failure criterion. Delamination propagation is governed by the energy release rates. After the occurrence of damage, constraint between sublaminates is removed to model the delamination and a ply discount model is used to account for the materialdegradation effects of matrix cracking. The establishment of a leakage path is comprised by jointed matrix cracking in each layer and delamination at each interface. Finite element analysis has been carried out to simulate the quasi-static indentation of filament-wound composite pipes. Despite the simplification of treating each damage mechanism independently (i.e., no direct interactions), good agreement has been achieved between experimentalresul ts and the predictions of the model for the load–indentation relationship, the evolution of multiple delaminations and the formation of a leakage path.

General expressions for energy–release rates for delamination in composite laminates

When two–dimensional or three–dimensional elasticity theory is applied to interfacial cracks, individual components of the energy–release rate are not well defined in their classical crack closure integral form due to oscillatory stress and displacement fields around crack tips. These are associated with physically inadmissible interpenetration of crack surfaces. When dealing with composite laminates, it is generally preferable to use a laminate theory. However, when applied to delaminations most laminate models can only produce a total energy–release rate but not the individual components of it. In this paper, expressions for individual energy–release rates have been derived for delaminations based on a sublaminate model. By considering delamination growth as a variational problem with variable endpoints, the total energy–release rate can be expressed simply in terms of stress resultant jumps and the derivatives of relative displacements between the delamination surfaces at its tip. The mode I, mode II and mode III components of the energy–release rate are then determined according to their definitions. The use of laminate theory eliminates oscillatory behaviour along with the stress singularity encountered in two–dimensional or three–dimensional linear elastic fracture mechanics theory for this problem. Instead, the effect of the singular stress field is reflected in the stress resultant jumps across the delamination tip. The individual energy–release rates derived in this way are all well defined. The present approach shows significant advantages over the popular finite–element–based virtual crack closure technique used for delamination problems.

Measurement of the critical energy release rate GIIC for filament wound GRP pipes

An experimental procedure is described for measuring the critical energy release rate GIIC of filament wound pipes by making use of lateral indentation experiments on these pipes and correlating the dissipated energy to the delamination area. Factors that might influence the results of this procedure are discussed. Two pipes of different nominal thicknesses have been tested and their critical energy release rates measured. The values obtained from different specimens of the same type show good consistency. The thinner pipes tend to produce a lower critical energy release rate. The measured values have been used as part of the input data in a numerical simulation of the problem of indentation of the pipes, which involves large deformations and extensive damage. Excellent agreement has been demonstrated between the results and experimental data, which serves as a validation of the procedure proposed for estimating the critical energy release rate.

Finite Element Models of the Human Shoulder Complex: A Review of Their Clinical Implications and Modelling Techniques

Summary The human shoulder is a complicated musculoskeletal structure and is a perfect compromise between mobility and stability. The objective of this paper is to provide a thorough review of previous finite element (FE) studies in biomechanics of the human shoulder complex. Those FE studies to investigate shoulder biomechanics have been reviewed according to the physiological and clinical problems addressed: glenohumeral joint stability, rotator cuff tears, joint capsular and labral defects and shoulder arthroplasty. The major findings, limitations, potential clinical applications and modelling techniques of those FE studies are critically discussed. The main challenges faced in order to accurately represent the realistic physiological functions of the shoulder mechanism in FE simulations involve (1) subject‐specific representation of the anisotropic nonhomogeneous material properties of the shoulder tissues in both healthy and pathological conditions; (2) definition of boundary and loading conditions based on individualised physiological data; (3) more comprehensive modelling describing the whole shoulder complex including appropriate three‐dimensional (3D) representation of all major shoulder hard tissues and soft tissues and their delicate interactions; (4) rigorous in vivo experimental validation of FE simulation results. Fully validated shoulder FE models would greatly enhance our understanding of the aetiology of shoulder disorders, and hence facilitate the development of more efficient clinical diagnoses, non‐surgical and surgical treatments, as well as shoulder orthotics and prosthetics.

State space approach with fem for the determination of structural behaviour of composite plates

Purpose An analytical investigation has been carried out for a simply supported rectangular plate with two different loading conditions by using 3D state space approach (SSA). Also, the accurate location of the neutral plane (N.P.) through the thickness of the plate can be identified: the N.P. is shifted away from the middle plane according to the loading condition. The paper aims to discuss these issues. Design/methodology/approach SSA and finite element method are used for the determination of structural behaviour of simply supported orthotropic composite plates under different types of loading. The numerical results from a finite element model developed in ABAQUS. Findings The effect of the plate thickness on displacements and stresses is described quantitatively. It is found that the N.P. of the plate, identified according to the values of the in-plane stresses through the thickness direction, is shifted away from the middle plane. Further investigation shows that the position of the N.P. is loading dependant. Originality/value This paper describe the effect of the plate thickness on displacements and stresses quantitatively by using an exact solution called SSA. Also, it is found that the N.P. of the plate, identified according to the values of the in-plane stresses through the thickness direction, is shifted away from the middle plane. Further investigation shows that the position of the N.P. is loading dependant.