Applied Mechanics

The following Sections are included:

• Preface

• Contents

Scale effects in mechanics are discussed focusing the concept of macroscopic aspects and microscopic aspects. Problems of scale effects in various fields are introduced. The mechanism of scale effects is analyzed and explained in each case. Non-dimensional method is introduced to meet the scale effect. Examples of scale effects are shown in the deformation and fracture of solids.

Sectional stability of thin-walled sections includes short and intermediate half-wavelength buckling modes commonly called local and distortional. Local modes involve only flexural deformations whereas distortional modes involve flexural and membrane deformations of the plates forming the cross-section and occur at longer half-wavelength. These modes may interact with each other or with long half-wavelength buckling modes such as flexural and flexural-torsional. Design methods to account for these modes in engineering design have been proposed and partially incorporated into current and pending design standards. The paper discusses these modes, their interactions and current and proposed design methods in modern design standards.

The realistic replication of mandibular motion for the construction of dental prostheses was the main purpose for studying mandibular kinematics in the past. Meanwhile, the research interest has shifted to the analysis of the function of the masticatory system, especially of the temporomandibular joint (TMJ). With the development of sophisticated microelectronics, mandibular tracking has been made possible in a precise way by recording all six degrees-of-freedom relative to the skull. The application of motion to craniomandibular anatomical models gives a non invasive dynamic insight into the TMJ. This virtual biomechanical technique (dynamic stereometry) facilitates dynamic, three-dimensional measurements within the joint. The present work deals with new methods developed for an integrated in vivo analysis of mandibular biomechanics. A series of studies were performed using these new methods. The aims of these studies were the animation of real TMJ morphologies with real jaw tracking data (which permits the motion characterization by means of time-dependent variables describing the relationship between the articulating surfaces), the quantitative analysis of the intraarticular distances in the TMJ related to functional mandibular movements and the use of the dynamic stereometry of the TMJ for the study of the joint load by means of finite element analysis.

In the last two decades, the steady interest in utilizing composite structures in offshore oil exploration and production has led to the conceptual birth of various designs with limited opportunities for actual implementation in service. Nevertheless, they are at the edge of full implementation as the certification, design, and manufacturing triad is optimized for economic payoff. To date, various research entities both in Europe and US have demonstrated the technically enabling features of composites in rigid and flexible tubulars, pipelines, pressure canisters, and topside grating. The present technical challenge lies in the integration of manufacturing process and material selection that will enable an accurate long-term response prediction of these heterogeneous and anisotropic structures. The logical step forward for offshore is to form alliances to define the service conditions and their impact on life of structures in deep-water applications without compromising the competitiveness in design and manufacturing.

Structural health monitoring (SHM) offers the potential for a quantum gain in efficiency for asset maintenance and structural integrity management. A successful implementation of SHM will need to integrate two distinct strands that have traditionally been pursued separately, viz. damage detection and damage assessment. Both strands require a sound understanding of damage mechanisms, failure modes, and of the possible contrast mechanisms that can be exploited for damage detection. The present talk presents recently developed theoretical concepts and results for detecting delamination damage in fibre-composite, plate-like structures. The modelling is based on the premise that the primary effect of a delamination is to reduce locally the plate's flexural stiffness. Accordingly, for the purposes of detection, the delaminated region can be treated as an inhomogeneity, having a lower (and generally complex) flexural stiffness. The interaction of plate waves with such inhomogeneities is discussed within the framework of Mindlin plate theory. This theory provides an accurate representation of the exact dispersion relation for the lowest antisymmetric mode of the Rayleigh-Lamb spectrum up to quite high frequencies, corresponding to wavelengths of three times the plate thickness, thereby providing an adequate and practical basis for resolution of the in-plane dimensions of a delaminated region. The wave interaction problem is solved explicitly in the Born approximation, which is appropriate for detecting relatively low levels of damage (barely visible impact damage). The issue of deriving a micro-mechanical estimate of the (complex) flexural stiffness corresponding to a given geometry of delamination is addressed separately…

In this paper the characterisation of jaw-mechanics by means of the finite screw and the wrench, is presented demonstrating its ability to provide a simplified, yet complete description of jaw function. With motion data and anatomical data from a single subject, the screw theory approach is implemented where the forces of the masticatory muscles are uniquely described at each instant by the wrench axis and its parameters. Incorporating the wrench axis description for muscle forces and moments during the motion with the finite screw representation of the motion, an integrated computer model of jaw mechanics is formulated. The results of an experimental program, involving Magnetic Resonance (MR) imaging for obtaining accurate anatomical representations of the masticatory muscles and the bony structures, as well as jaw-motion data collected via an optoelectronic recording instrument, are presented. The results show that the experimentally derived, screw-based model conforms to the clinically observed behaviour of jaw function and can provide a quantitatively accurate means of analyzing mandibular biomechanics.

A spiral position of a cochlear implant electrode, closer to the nerve cells in the center of the cochlea, should result in reduced threshold currents and increased dynamic range of auditory stimulation. Patient psychophysics studies using various electrode configurations have confirmed these hypotheses. An intracochlear electrode array has been developed for safe yet simple implantation, resulting in peri-modiolar placement, close to the auditory nerve cells. Key components of this development have been the study of the biomechanics of cochlear electrode insertion, concept design and evaluation of alternatives, and validation of the preferred design choice. The new electrode array is made pre-curved, conforming to the inner curvature of the average cochlea after insertion. This design concept is based on a stylet introducer, which holds it almost straight and imparts the assembly appropriate properties for insertion. This paper explains the biomechanics of the insertion of the new electrode array, with emphasis on comparisons with other existing or considered cochlear electrode designs. Biological safety aspects of the validation of this electrode, leading to its regulatory approval and product launch, are also presented.

This study examined static compression properties of sheep eyes and dynamic forces on the eye due to impact of squash and cricket balls with the eye. Dynamic forces were measured by using an accelerometer attached to the non-impact side of an impacting ball and dropping the ball onto the eye from two-drop heights. Bio-dynamicists may wish to employ this data. in their models for eye injury.

An investigation into the surgical instrument/tissue interface of 2-Dimensional and 3-Dimensional wave profiled surfaces. Pairs of aluminium-profiled blocks were machined with teeth of differing size and shape. These were used to grip leather samples under a constant pneumatic pressure. The sample was extracted at a velocity of 50 mm/min until either the grip or the sample failed. The load at which the grip failed, the maximum load generated, and the type of failure was recorded. A quantitative relationship between wave amplitude and release load was established.

Stress fractures, which are a common injury in athletes and the military, are a result of applying repetitive cyclic loading to bone. They are commonly diagnosed in the tibia, either at the midshaft or at the junction between the mid and distal thirds of the tibia. Although much is known clinically on the diagnosis and etiology of stress fractures, the mechanical factors that contribute to their development are not as well understood.

Using a gait analysis lab with a full marker system, values of the forces and limb positions were obtained from an aerobic athlete running across the plate. These values were used in conjunction with supplementary information obtained from cross-sectional CT scans taken at regular intervals along the athlete's tibia to perform a mechanical analysis of the bone. The objective of this analysis was to determine the location and approximate magnitude of the highest stresses and which load or combination of loads produced them. This paper will discuss the results generated from the mechanical analysis and their clinical significance. In addition, the application of the gait data to a Finite Element (FE) Model will be discussed in the conference presentation.

Measurement of bone density for women is of great importance especially for certain age, as by aging bone density and percentage of bone mineral content decrease causing weakness in bones. There are several techniques for bone density measurements but the most important one is the ultrasound waves, which are widely used, in medical applications. When voltage is applied, high frequency longitudinal waves are produced and measurements are made of the ultrasonic attenuation. In this research work, samples are taken and positioned between. two broadband ultrasonic transducers, one of them transmits short pulses of broadband ultrasound and the other used as a receiver, and a relation-ship between ultrasonic attenuation and frequency is established. In this paper, the relationship between attenuation coefficient and bone density is determined. Also, a correlation between bone density and percentage of mineral content is established.

A general method of constructing theoretical models describing the compaction and shearing of sands, soils, powders and other granular materials is described. The procedure is based upon the laws of thermo-mechanics, so that the models automatically obey the Second Law inequalities, and there is no need to appeal to "quasi-thermodynamic" stability postulates. For illustration the theory is couched in terms of the variables commonly used for tri-axial tests. It is shown that existing models can be made more realistic in a systematic way by modifying the initial dissipation function to include non-associated flow rules and shear as well as volumetric hardening. This new approach to model-building also pin-points some shortcomings of some existing procedures.

A nonquadratic yield function (Yld96) for aluminium alloy sheet, which simultaneously accounts for the anisotropy of uniaxial yield stress and r-values, was developed by Barlat et al.¹. This model was based on isotropic hardening and was recently extended to account for the difference between tensile and compressive yield strength², resulting in predicting earing profiles in good agreement with experimental results. Isotropic hardening, however, is mainly applicable to monotonic loading. For problems with stress reversal, it fails to predict the material behaviour due to the Bauschinger effect. In this paper, an elastic-plastic tensor of the tangent modulus for plane stress problems based on Yld96 is derived, which includes the isotropic and kinematic hardening to account for the Bauschinger effect. It provides a convenient means for numerical analysis of plasticity such as the finite element analysis.

Mohr-Coulomb Theory (MCT) assumes that the normal to the plane of failure lies in the plane containing the maximum and minimum principal stresses. This assumption can be derived from Elasto-Plasticity Theory (EPT). Both MCT and EPT predict the same orientation of the failure planes for associative failure theories, but this orientation is independent of the elastic parameters in EPT. In this paper we present a new (second-order) Rigid-Plastic Theory (RPT) of failure and show that it also agrees with EPT for associative failure theories, but it can predict failure angles different from those predicted by EPT for non-associative failure theories. The experimental results on the orientation of failure planes in unconsolidated sands are consistent with the predictions of RPT.

Image measurement based on the Fast Fourier Transform convolution theorem has been combined with a finite difference algorithm to produce image tensor analysis (ITA), a numerical tool for experimental determination of deformation for historic continua.

ITA allows the deformation within an image-compatible material body to be quantified using digital video footage taken from experimental events. Hence the deformation output from computational modelling of the same event can be directly compared to the measured deformation field given by the image-based numerical solution.

The current research application of ITA aims to provide experimental deformation data and characterisation of the constitutive behaviour of the fibro-porous media prepared sugar cane. This paper outlines the numerical procedure of ITA and details the possible post processing applications with respect to both data acquisition and constitutive modelling.

Thermal fracturing (spallation) is a method of breaking rocks and concrete. It is based on heating the surface (usually with open flame), utilising thermal stresses to develop small chips (spalls) at the surface. The method is cheap and easy to use, however, its practical applications are restricted by the relatively low pace of the breakage and by sensitivity to the material properties and microstructure. Thermal spallation only occurs when the temperature is kept just below the melting point of the particular material. The process is also very sensitive to the presence of water in materials such as rock and concrete. It has been found that thermal fracturing is controlled by the internal permeability of the material in materials with low permeability, water in the pores and pre-existing cracks cannot escape during heating, evaporates and drives the crack in uncontrollable manner. In this case, the sample fails in splitting without any spallation. The difference between these two failure modes is also controlled by the material's tensile strength.

By adopting Kane and Mindlin's assumption that the through-the-thickness extensional strain is uniform through the plate thickness, a generalised plane-strain theory is developed for transversely isotropic plates. The three-dimensional governing equations are successfully reduced to two coupled equations in the two-dimensional space. With the new theory, explicit solutions of the three-dimensional stresses, especially the through-the-thickness component, around a circular hole and a U-shaped notch in a transversely isotropic plate of arbitrary thickness are derived. The analytical solutions are verified by comparing with finite element results.

This paper presents a new and relatively simple engineering method for calculating the stress intensity factors for small-to-medium cracks emanating from a notch under arbitrary loading. The formulation can be used in calculating the fatigue life of notched components as well as in the shape optimisation problems with durability constraints. Several examples are considered to demonstrate the advantages of the present method in comparison with both existing approximate approaches and finite element techniques.

The fatigue design of tubular nodal joints in the latest IIW and CIDECT fatigue design guidelines is based on the hot spot stress method. Stress concentration factors (SCFs) form a significant part of the design of welded connections using the hot spot stress method. This paper describes the determination of experimental stress concentration factors in thin-walled tube-to-tube T-joints made up of square hollow sections, due to in-plane bending moment in the brace. The effect of induced axial load and induced bending moment in the chord on experimental SCFs is studied. The experimental SCFs are compared to the SCFs derived from existing parametric equations. It is found that the experimental SCFs are significantly lower than the SCFs from existing parametric equations. Observations are made in this paper which form a preliminary part of future research into SCFs of thin-walled tubular connections.

Shape reworking has historically been used to remove cracks and reduce the peak stress at critical locations in the F-111 in service with the RAAF. Optimal reworks for the most critical fuel flow vent hole fatigue locations in the F-111 wing pivot fitting have been determined using a recently-developed finite-element-based gradient-less shape optimisation procedure, and typically provide a 35% reduction in peak elastic stresses as compared to current rework shapes. These optimal rework shapes were developed using both 2D PAFEC and 3D MSC.NASTRAN finite-element models. The final robust optimal rework shapes provide a low peak stress over a range of conditions. The numerical strain results have been validated with a recent full-scale static wing test. It is anticipated that the stress reductions achieved will provide a basis for extending recommended inspection intervals, thus reducing costs and increasing aircraft availability until the planned withdrawal date of 2020. It is planned that the RAAF will commence modifying F-111 fleet wings in 2001.

Repeated thermal shock loading is common in many industrial situations including the operation of pressure equipment found in thermal power stations. Thermal shock can produce a very high stress level near the exposed surface that eventually may lead to crack nucleation. Further crack growth under the influence of repeated thermal shock is a very complex phenomenon due to both the transient nature of the highly non-linear thermal stresses and the strong influence of the environment. This paper describes an experimental analysis of crack growth in heated carbon steel specimens exposed to repeated thermal shocks using cold water. Analysis of the effect of steady state primary loads on the growth of the cracks is isolated using a unique test rig design. Environmental effects due to the aqueous nature of the testing environment are found to be a major contributor to the crack growth kinetics.

Monotonic load tests of plain and notched cylindrical tensile specimens have been made using a variety of aluminium, brass, and steel materials. It is shown, as would be expected, that many of these normally ductile materials may exhibit brittle fracture in notched bar tests, depending on the notch geometry. The apparent notch toughening effect was measured as the ratio of nominal breaking stress of notched versus un-notched specimens. Notch toughening is then used as a parameter to describe the relative ductility or brittleness of a particular fracture, given that the material is normally ductile. Elastic-plastic analyses of the two types of notched specimens were carried out using highly detailed finite element modelling, to better understand the mechanics of deformation processes in the regions of high plastic strain. Good agreement was achieved between numerical analyses and experiments, with finite element analysis replicating the notch-toughening behaviour found in load tests. These results are used to demonstrate that the continuum plasticity model consistently predicts monotonic failure curves (applied load versus axial deformation) for notched metallic specimens, whether the failure is brittle or ductile. The ductility or brittleness of the failure mode are shown to depend on notch geometry and material characteristics, including yield strength, hardening exponent, and fracture ductility of a plain bar in uniaxial tension.

A numerical analysis was conducted, using previously obtained experimental results, to establish basic toughening mechanisms and fracture behaviour of an interlayer-toughened composite material, containing particulate modified toughened layers. The aim of the analysis was to examine the interaction between the particles and the crack tip during crack propagation. Good agreement with the experimental data was found.

A novel mechanical surface treatment using laser-induced shock waves has been applied to a 7xxx series aerospace aluminium alloy to improve the fatigue life. Laser shock peening (LSP) is applied by using a high energy pulsed laser to create a high amplitude stress wave or shock wave on the surface to be treated. This stress wave propagates into the material, causing the surface layer to yield and plastically deform, and developing a residual compressive stress. The residual compressive stress from laser peening extends much deeper than that from shot peening. The fatigue behaviour has been investigated at different applied stresses and is compared with those of conventionally shot peened specimens. The fatigue life of laser-shocked specimens was found to be higher than that following shot peening, particularly at extremely high stress (450MPa). This large increase in fatigue life appeared to be the result of the deeper residual compressive stress and the smoother surface produced in the laser treated specimens. For this aluminium alloy, the laser shock surface treatment does not appear to create any microhardness modifications. Therefore, hardening does not appear to be responsible for the fatigue life extension.

An experimental compliance method and a three-dimensional acoustic emission source location method were simultaneously used to determine the location of the crack tip in fracture toughness tests of Kimachi sandstone. Tests were done using the semi-circular bend specimen under three-point loading. Both methods gave agreeable results for the advancing crack length and they were used to construct the R-curve. The method enables the determination of size-independent fracture toughness of the rock material using a single specimen.

The influence of adhesive reinforcement on the Mode II fracture toughness of an end notch flexure (ENF) specimen is investigated with a computer model based on a variational theorem that has been shown to accurately predict stress fields and energy release rates for cracked laminates. A large radius axisymmetric fracture mechanics model was developed and used to determine the energy release rates for midplane cracking in aluminum specimens and in aluminum bonded specimens with varying amounts adhesive reinforcement. The energy release rates, which were determined from the present model, were used to determine the fracture toughness for specimens in a systematic experimental program and were compared to closed form analyses from the literature. The amount of reinforcement in the adhesive layer has been shown to have a significant effect on the Mode II adhesive fracture toughness and cracking behavior.

In the present study, a boundary layer model was proposed to analyze the effect of constraint parameters on mixed mode fracture toughness. A new definition of crack-tip constraint parameter, Q was presented to establish a unique relationship between the fracture toughness and the constraint parameter developing finite element models. The T-stress was also related to Q considering the far field mixity parameters employing modified boundary layer analyses.

Crack growth prediction of components subjected to spectrum loadings is an important issue in fatigue life assessment of industrial components. Effects of load interaction cause acceleration and/or retardation in fatigue crack growth rate. Disregarding of this phenomenon may lead to the nonconservative life prediction in some cases. In this paper two methods for load interaction consideration are used. The first model is Chang-Willenborg that is a modified Willenborg model and the second one is a new model based on the compressive residual stresses at the crack tip and original Willenborg model. In the development of the new model the required residual stresses at the crack tip are obtained from an elastic-plastic finite element analysis. Also the Original Rainflow Counting Method and Modified Rainflow Counting Method have been used to prepare the spectrum loading data. Using the above mentioned cycle counting methods and load interaction models, the crack growth of the plates are obtained for two different flight spectrum loading. Finally the results obtained from different methods are compared with those obtained from experiments presented by the others.

An approach for structural damage diagnosis employing the wavelet transform technique is proposed in the paper. A damage identification system with built-in piezoceramic transducer is developed in correlation with a stress wave propagation model. In the system, signal-processing techniques in the time-frequency domain are applied to reduce noise interference and facilitate the acquisition of diagnostic information from acquired raw data via signal processors. In addition, the capability of this system is investigated through experiments considering a beam pre-treated with various degrees of damage, which aims at simulating the damage growth. It is shown that the damage position and degree in a flexural beam can be detected quantitatively and efficiently by this approach.

Preliminary work has shown that damage in a short round bar can be quantified by determining the attenuation of an ultrasonic stress wave. A bar provides a waveguide for the impulse to travel and has a simple geometry. A shoulder or other geometric feature will reflect the excitation, which would change the recorded response. To investigate the possibility of applying this technique in situ, three bolts were also loaded under conditions of varying levels of uniaxial tension, in a large steel block. The aim being to observe the effect of an alternative load path and a mean stress on the attenuation of an acoustic stress wave and to compare this with an untensioned bolt. A linear relationship between attenuation and crack depth was observed.

The recent application of x-ray computed tomography has enabled new 2-D and 3-D in situ diagnostic and visualization capabilities in both the pre- and post-mortem damage assessment of impacted ceramic targets. Examples of such capabilities presented include: (a) 2-D and 3-D asymmetrical mesocracking in TiC, (b) residual tungsten alloy penetrator distribution in TiB₂, (c) subsurface cracking in body armor ceramic panels and (d) cracking of monolithic ceramic tiles contained within a cast or welded and hot isostatic pressed metal encapsulation. These results have enabled, and will continue to provide, damage descriptions that are far more accurate and revealing than previous approaches and that affect more efficient and detailed visualization and modeling of the various complexities present in practical armors. This innovative nondestructive CT technique provides a vital ballistic damage assessment approach which has applicability to other physical damage situations such as quasi-static fracture, fatigue and corrosion as well.

Expanded Polystyrene (EPS) geofoam is a thermoplastic material (of synthetic origin) used increasingly as an ultra-lightweight fill in geotechnical applications. Its use extends from replacing the low load bearing, highly compressible subgrade soil in road pavements to as a backfill body in the construction of retaining walls and embankments. In these and other sub-surface constructions, the material is subjected to varying moisture conditions. It has been observed in field studies that as much as 9 % moisture pickup by volume occurred in blocks lying permanently under the ground level for 9 years. Moisture impinges on geotechnical design in that it not only increases the unit weight of the EPS geofoam, but may significantly affect the mechanical properties as well. As little by way of experimental evidence is available to form appropriate conclusions, the authors have undertaken a study to investigate moisture effects on the mechanical behaviour of EPS. In the study, EPS samples were prepared and carefully soaked in distilled water for various lengths of time ranging from two weeks to one year. At appropriate points in time, batches of thirty soaked samples were withdrawn and subjected to uniaxial and triaxial compression tests to assess the mechanical behaviour. Preliminary findings show that the volumetric moisture content (VMC) increases with the duration of soaking. It appears that moisture has little effect on the material elastic stiffness or its yield stress, but causes an increase in the plastic modulus.

Semi-analytical solutions of free strain consolidation of fine-grained soils by vertical drains are developed in this paper. The solutions obtained are semi-analytical in the sense that the problem is first solved analytically in the Laplace transform domain and the corresponding values in real time and space are then obtained by numerical inversion of the Laplace transform solutions. Solutions are obtained for the distribution of excess pore pressure and degree of consolidation subjected to a single step incremental loading.

Mobile fleet of mining equipment is one of the highest cost contributors in a mining operation. This paper describes various advancements of haulage methods that reduce the unit cost while improving safety of mining. Proper maintenance practices are of paramount importance to improve the reliability and the availability of equipment. The performance analysis of a few mines is given in relation to key performance indicators that describe the availability and the utilisation of equipment. Some benchmarks for equipment performance based on achievable targets are suggested and compared with the actual performance.

This paper provides an overview of the energy and crush space required to provide survivable impacts between motorcars and energy-absorbing poles. In addition this paper considers some of the constraints that would apply if those poles or posts would be required to be remaining upstanding and serviceable after the crash. A simplified acceleration/displacement pulse for the combined car and pole is used in all the modeling described here. This pulse is arrived at using results from full-scale tests and FEA simulations for impacts from 60 km/h and for a maximum acceleration of 20 g. These limitations and the shape of the pulse give us the total crush depth for any given collision energy. The depth into the car, at the point when 20 g is reached is arrived at using published analysis of full-scale car-to-pole tests and it is a function of the car's mass, width and length, and the pole diameter. The division of crush depth between the pole and car is then estimated using the assumption that the pole can be designed to crush from the time that the car reaches 20 g acceleration. The results show that the poles that bring about 20 g acceleration in the car can be provided only for a narrow range of car masses. A pole that suits medium sized cars will cause reduced acceleration to heavier cars, to the point that it may be overrun and demolishing by much heavier cars. The same pole will cause higher accelerations for cars that are lighter than the chosen mass, to the point that it may be equivalent to a rigid pole for very light cars.

In this paper an analytic equation for dynamic shear stress in an orthotropic hollow cylinder under torsional impact loading is derived using the eigen-function method. Mixed boundary conditions are assumed. Example calculations are presented applying this equation to determine stress time history response and distribution for two different orthotropic thick-walled cylinders under torsional impact.

A theoretical analysis of an axially compressed circular metal tube deforming in progressive axisymmetric folds is presented. This new model is developed based on our experiments and finite-element analysis. It helps explains our previous empirical finding that the non-dimensionalised mean crushing force P_m/M₀ is not proportional to (D/h)^0.5. An empirical formulae for the mean crushing force and half wavelength are established to best fit the numerical calculations. Comprehensive comparisons have been made among the present theoretical prediction, experiments and FEA results.

Titanium alloy Ti-6Al-4V, produced via a new low-cost Electron-Beam Cold-Hearth Single-Melt Process, was ballistic limit tested against 12.7 mm projectiles. A brief overview is given of emergence of low-cost titanium as a real option for armour applications as well as the characteristics of the tested Electron-Beam Single-Melt Titanium. Simple plasticity-based analytical models are used to predict the ballistic limit for the non-deforming projectile case and to illustrate how important it is to understand the detail of target deformation and failure. The work provides essential background to future studies that will better characterise the ballistic deformation and damage in the target plate to further improve predictive modelling efforts.

In the present analysis harmonic extensional wave propagation in a pre-stressed, incompressible, symmetric layered composite is considered. The bi-material composite consists of two isotropic elastic materials. The imperfect interface is simulated by a shear spring type resistance model that can accommodate the extreme cases of perfect bonding or a sliding interface. Using the technique of the propagator matrix, the dispersion relation is obtained. Four possible cases of dispersion curves that depend on the limiting shear wave speeds of the bi-material composite are discussed. The effect of the imperfect interface is more pronounced for low wave numbers.

This study examined mechanical and dynamic properties of a range of balls used in many sports with a view towards relating these properties to physical injury due to impact of a ball and a person. The study examined impact forces, force durations, impulse, impact pressure and rebound heights for thirty different balls that included fourteen cricket balls. The data presented may be employed in models for ball-human impact studies.

A case has been encountered in a critical helicopter gearbox, where a spalling fault in a bearing inner race manifested itself almost exclusively at discrete frequencies with a spacing equal to the speed of the shaft on which the bearing was mounted. A fault on a gear on the same shaft would produce the same pattern of frequencies, and so a method is sought to differentiate between the two situations. In the specific case of a fault in a bearing inner race, which passes through the load zone at a rate equal to the shaft speed, the signal is not periodic, but cyclostationary, because the rolling elements are in a different position on the rough fault surface for every revolution. Cyclostationary signals have the property that their autocorrelation function is periodic, so that their spectral correlation contains lines at discrete cyclic frequencies (α), even though it varies continuously in the normal frequency (f) direction. The power spectrum is represented by the line α = 0, and the spectral correlation of stationary signals is zero for all other α. The spectral correlation of periodic signals contains discrete peaks in both directions. The proposed method thus uses the fact that at values of α representing multiples of the shaft speed (thus eliminating additive stationary noise), the spectral correlation will have non-zero values between the peaks that come from the periodic part of the combined signal. The paper illustrates the technique using both simulated and actual signals.

In the free axisymmetric blowing of hollow axisymmetric forms, a stable anuerism often develops which then travels through the form. This paper presents a procedure for carrying out experiments where by stresses and strains can be measured during axisymmetric free blowing. This paper explores the relationship between material properties and this behaviour to explain the conditions under which stable anuerisms can occur.

At the University of New South Wales a machine vision system is being developed for creep strain measurement at room and elevated temperatures. The system is based on a digital speckle correlation technique and it uses white light. This paper describes the development of this system and its successful application to measure creep strain at room temperature using a specimen made of nylon 66.

Pressure sensitivity adhesive tapes are become more common in many environmentally demanding applications. One such example is the use of paint film appliqués that have pressure sensitive adhesive backings applied to aircraft as paint schemes. A concern that exists with such paint scheme applications is durability of the adhesive and adhesion interface in the harsh environment of flight. A holographic interferometry study has been undertaken to study the effects of adhesive degradation and surface condition deterioration. The results show that poor surface conditions and bondline defects can be identified in holographic interferograms. Characterisation of several forms of bondline defects are presented and qualitatively analysed.

This study aims to evaluate the machinability of polymers and understand the interaction between strain rate and temperature rise during orthogonal cutting. The materials used were high density polyethylene (HDPE) and low density polyethylene (LDPE). It was found that the average surface roughness and shear strength of both HDPE and LDPE decreases with increasing the cutting speed, demonstrating that the strain rate has a greater influence than that of the temperature rise. A relatively high mechanical strength is preferred in order to create a smoother surface of soft polymers with a low glass transition temperature and mechanical strength. It was revealed that the chip deformation mechanism varies with material properties.

Delamination is one of the serious concerns during drilling. This paper presents a comprehensive analysis of delamination in use of core drill. The critical thrust force at the onset of delamination is predicted and compared with the twist drill. The well known advantage of industrial use of core drill is fundamentally illustrated. A guide for drill design can be developed based on the proposed models, and this approach can be extended to examine the effects of various drill or future innovative drill design.

This paper studies a method of improving the tribological and mechanical properties of steel elements by grinding. It was found that a proper use of grinding heat and stress can create favorable microstructures and promote high wear and fatigue resistance. The thickness of the treated surface layer can be up to 1mm. An enhanced phase transformation, intensive dislocations and a uniform carbon distribution make the microstructure of the layer beneficial.

This study tends to investigate the reliability of the surface modification by grinding through understanding the heat flux formation and temperature distribution in the process. A method of capturing the heating zone movement by means of an optical microscope and a CCD camera was used to detect the temperature rise in the subsurface of a component subjected to grinding. The heating parameters, temperature field and stability of the method were evaluated. It was shown that a triangular heat flux gives the best fit for up-grinding. The contact grinding length can deviate significantly from the geometrical contact length due to thermal expansion of workpieces. It was also proved that stable grinding conditions might produce beneficial surface layers in ground component.

Various mechanical surface enhancement technologies have been invented and developed to induce a protective layer of compressive residual stress at the surface of aircraft engine components where the operating loads are tensile dominated. The benefits of these surface enhancements are to prevent crack initiation, to retard propagation of small cracks and even to resist corrosion and wear damage. In this paper the literature on these technologies and their effects on component reliability and durability is reviewed, with an emphasis on Shot Peening (SP), Laser Shock Peening (LSP) and Low Plasticity Burnishing (LPB). The relative advantages of these three surface enhancement technologies as well as their limitations are identified and evaluated. The most important issues for these three technologies, and the relative merits of the resultant residual stress fields are presented and discussed with a view to enhancing engine component operating lives.

The deformation zone in cold rolling includes elastic entry, elastic exit and plastic deformation regions. The elastic deformation zones have an effect on the lubricating film thickness and the accuracy of the final product. Based on the ABAQUS/Explicit code, the authors analyze cold strip rolling, and focus on modelling of the elastic entry region of a roll bite. The deformed elastic entry zone is obtained from the FE simulation and the effects of tension and rolling speed on the deformation are studied. The calculated rolling forces are close to the measured values. This model can predict the hump at the elastic entry zone of the strip, and contribute to the accuracy of the final product through a correct roll gap set up and improvement of the mixed film lubrication.

The threading and acceleration processes of tandem cold rolling are modeled and dynamically simulated. Fluctuations in the processes are investigated theoretically and taken into account in the simulation, such as motor speed droop; speed dependent parameters during acceleration including friction coefficient between the rolls and strip, and oil film thickness in hydro-dynamically lubricated roll neck bearings. Strip thickness deviations under different threading conditions are compared and optimal threading speed is searched based on a heuristic optimization procedure. For acceleration process, the compensation coefficient of Automatic Gauge Control (AGC) is optimized to reduce off-gauge length of the strip. The simulation results correspond well with plant data and show that the off-gauge situation with optimal threading and acceleration parameters applied can be alleviated.

A thermal model has been developed for mixed-film lubrication in cold rolling. The energy equation has been used to determine the temperature variation in the inlet, outlet and the plastic work zones. The temperature at the asperity contacts in the plastic work zone is also calculated. The influence of the thermal effect on the film thickness and pressure is significant and should be considered in a mixed film model. The model has been validated by experiment.

This paper presents an elastic-plastic finite element model for asperity deformation in cold rolling. The asperity is assumed to be spherical and an isotropic hardening material is applied. The emphasis of this simulation is placed on the relationship between contact pressure and actual area of contact. The influence of friction is investigated as well. Results and discussion of variations of main contact parameters are presented. Comparison with previous work shows that most results are in good agreement.

Continuously variable crown (CVC) system is an effective method for controlling strip flatness and profile. The key issue in the development of the CVC system is the design of the roll profile, which in this paper is based on a cubic function. Not only the desired variable crown can be obtained, but also the axial force acted on the rolls can be minimized.

In this paper, a finite element model was used to determine the transient temperature field of the transverse section of strip in the whole hot strip mill production line. The simulation shows that the calculated values are in good agreement with the measured values obtained from a hot strip mill. The temperature field across the thickness becomes uniform at the exit of finishing stands, but it shows that there is about 100 °C difference between the center and the edge of the strip. Before the strip enters the finishing stands, a temperature drop of 110 °C between the head and tail of strip has been found. However, the temperature drop reduces to 10 °C as the strip is processed at an accelerated speed of 0.14m/s² through the finishing stands.

The safety of the universal joint plays a key role in a plate rolling mill. The spindle angles and friction have significant effects on the stress states of jaw and jaw coupling of the universal joint. In this paper, the authors considered the additional bending moment due to the spindle angles of the joint and the bending moment due to the friction between the bearing and the jaw and its couple respectively in addition to the torque effect from the main drive. The equivalent stress and the Tresca stress have been analyzed with the ALGOR Finite Element Analysis System. The results show that the higher is the friction coefficient, the greater the influence of the friction moment plays on the jaw couple, while the smaller the friction coefficient, the greater the influence of the additional bending moment on the jaw couple. The friction moment has been found to have a greater influence on the stress state of the jaw couple than the jaw. These were verified by the failure of the work roll at the coupling end.

In this paper a new computing algorithm for the rigid-plastic finite element method is presented. The iteration method is used to solve the non-linear stiffness equations, and the error generated from the linearization of Newton Raphson Method is removed. Numerical tests show that the computing algorithm has a good convergence, and the same convergence results can be obtained for different initial guesses. The computing algorithm has shown that it is suitable for the simulation of rolling process.

In both rolling theory and practice, the important factor that must be considered is friction on the surface of the roll-strip. In order to improve the accuracy of on-line controlling models for hot strip rolling, in this paper, the rigid plastic finite element method is employed, considering constant friction, friction variation and mixed friction models to simulate this strip rolling. The results of simulation, such as rolling pressure and forward slip, taking into account the friction variation with suitable models in the roll bite are in good agreement with the experimental values. The modelling of friction variation with suitable models and mixed friction in the roll bite is stable and applicable, which can improve the accuracy of on-line control models.

A semi-continuous extrusion technique for the manufacture of metal tubes has been developed recently: In this method a tubular billet gripped in a urethane clamp is advanced repeatedly to extrude through a die. A machine utilizing this principle has been built and tested. The extrusion cycle consists of clamping the billet; advancing it to extrude a pre-determined length; releasing the clamping force (freeing the tube to stay onto the die); and, finally retracting the clamp to its initial position. A mandrel sliding inside the billet prevents it from collapsing under high radial pressure caused by clamping. The clamped billet, gripping the mandrel, is moved toward the extrusion die. Extrusion is stopped before the front-end of the mandrel reaches the deforming metal. At the end of each extrusion cycle the mandrel is free to slide, and it is retracted by a spring attached to its back-end, the spring being stretched by the mandrel moving forward with the billet during extrusion. Results from the investigations conducted are presented in this paper.

In this paper we investigate the use of Tchebychev nets to determine shear strains that occur in the mapping of flat blanks in sheet forming processes to free form surface shapes with Gaussain curvature. A draping algorithm has been adapted which deforms an inextensible net over a desired forming shape. Such a net deforms through shearing of the elements. However, there is an overall loss of area. Therefore, techniques of adjusting it to maintain continuity of area are presented. Such a modified net is able to give reasonable predictions of strain and is appropriate for assessing formability of a part early in the design process. A demonstration system based on CAD based NURBS surfaces has been generated for comparison to experimental results.

Shear and longitudinal mode ultrasonic wave speeds have been measured in powders of copper, aluminium and stainless steel during compaction. At the outset, all powders were nominally spherical, providing similar pore shapes. The wave speed measurements were used to calculate effective elastic moduli (Young's modulus, Poisson's ratio) for the powders, and monitor their evolution with compact density. Effective Young's modulus at a given density was found to be different for each material, though all increased monotonically with density in a qualitatively similar fashion. Normalisation of the measured Young's modulus by the respective solid phase value for each particle material accounted for the quantitative differences in the density dependence. Effective Poisson's ratio had a concave-shaped dependence on density for all powders, which was quantitatively similar over the range of densities studied. These results indicate that for compacts with equivalent pore shapes, the dependence of Young's modulus on density scales with the value of the solid phase Young's modulus. In contrast, the effective Poisson's ratio of the compact showed no material dependence, indicating it may be due primarily to the shape of pores in the compact.

The effect of particle size on the dynamic strength of liquid-bound granular materials was investigated. Four different size distributions of glass ballotini were used. Three narrowly sized fractions with mass median sizes of 28, 48 and 90 µm and one broad size distribution with a mass median size 71 µm, but a surface-mean size of only 35 µm. Pellets of length 25 mm, diameter 20 mm, porosity 35% and liquid saturation 70% were deformed at speeds varying from 0.03 to 150 mm/s. Water, glycerol and various silicone oils were used as the liquid binder. The strength of the pellets was strain rate dependent. The strength measured at low strain rates was not necessarily indicative, even qualitatively, of the relative strength ranking of different formulations at high strain rates. The peak flow stress (σ_pk) versus strain rate data was found to collapse onto a single curve when plotted in terms of two dimensionless groups.

The equations of motion for finite macrocsopic deformations of an idealised, monodisperse granular material are obtained by considering a statistically representative particle. The discrete equations of motion of the reference particle are homogenized by differential expansion of the difference expressions of the discrete model. We derive objective constitutive relationships relating the Jaumann rates of stress and moment stress to the relative strain- and curvature rates respectively. The elastic moduli are expressed in terms of the contact stiffnesses, solid volume fraction, coordination number and diameter of the sphere.

In this paper, an experimental method is presented to measure the properties of slug (an aerated material bed). Based on the measured properties, pressure drop in low-velocity slug-flow pneumatic conveying can be predicted accurately. The method can be applied to bulk solid materials with regular, irregular and/or unusual physical properties (e.g. different shape, density, size and size distribution), as long as they are good candidates for this mode of pneumatic conveying.

This paper presents research into the development of a computational system for modelling of complex particulate flow problems using the discrete element method (DEM). Important features of the system are: 1) particle generation with user-specified particle angularity and distribution in two and three dimensions; 2) a range of adhesion laws to enable clumping of particles and subsequent fracture / floculation based on a user-defined tensile force; 3) specification of interparticle and particle / wall friction and cohesion; 4) general prescription of drag kinematics; 6) solution efficiency; 7) post processing involving user defined cutting planes and animation.

With increasing computational power DEM is becoming a tool which can be used to explore complex modelling processes in three dimensions on a desktop computer. Results are given for two and three dimensional modelling of industrial flows: dragline bucket filling; flow from bottom dump rail wagons and silo flow.

This paper demonstrates peeling effects on performances of piezoelectric actuators and sensors in a smart beam with symmetrically bonded piezoelectric patches. The numerical results including the peel stress are presented and compared with those of the shear stress model. Interacting shear and peel stresses induced by the applied voltage, the actuated forces and displacements in the host beam are calculated. The effects of the actuated peel stress on the performance of the host beam are studied. The transferring shear and peel stresses, and sensing electric charges caused by a bending moment are also calculated and compared with those of the shear stress model, and then the effects of the sensed peel stress and the sensing charge are investigated. The results show that the peeling effects on the smart beams with piezoelectric patches are significant.

This paper reports on the application of multiple integral simulation tools to determine the characteristics of structural wave ultrasonic non-destructive evaluation systems for isotropic and anisotropic materials using adhesively bonded piezoelectric transducers. The electro-mechanical transduction behaviour of circular, rectangular and interdigital piezoelectric transmitter and receiver elements is discussed. The results demonstrate the importance of the application of advanced simulation tools to gain a better understanding of the complex electro-mechanical transduction processes and transducer-structure interactions, which is essential to enable the development of reliable quantitative non-destructive evaluation and health monitoring systems. In addition, it is shown that a hybrid numerical-experimental methodology is an efficient approach to optimise the performance of the electro-mechanical components of an acousto ultrasonic measurement system for a specific application and an important tool in the validation process of improved models.

Shape control of a smart structure is deemed as achieving the desired shape of the structure through optimizing the electric fields applied to and the loci and sizes of the piezoelectric actuators attached to the structure. In this paper, a finite element analysis software has been developed for analyzing static deformation of smart composite plate structures with non-rectangular shaped PZT patches as actuators. The mechanical deformation is modeled using a 3^rd order plate theory, while a layer wise theory is used to model the electric field. Numerical examples are obtained for the optimum values of electric fields in actuators to achieve the desired shapes using recently developed Linear Least Square.

The debonding between the piezoelectric actuator/sensor layers and the host structures can result in significant changes to both open loop and closed loop properties. To investigate the effects of the debonding of the piezoelectric actuator layer from the host beam on its dynamic behavior, an analytical model of a beam with debonded piezoelectric layer is presented. In addition to the host beam and the piezoelectric layer, the adhesive layer is also considered. In the perfect bonding region, the adhesive layer is modeled to carry uniform peel and shear stresses throughout its thickness, while in the debonding area, it does not transfer any peel and shear stresses between the host beam and the actuator layer. Both displacement and force continuity are imposed at the interfaces between the debonded and perfectly bonded regions. Based on this model, a semi-analytical solution is given to obtain the frequency response of the beam with debonded piezoelectric actuator layer by employing integration transformation, and the debonding effects on the frequency response of the beam can be investigated.

This paper reports on a preliminary experimental study in which an array of surface mounted lead zirconate titanate elements (PZT) are used for the in-situ detection of disbond growth in a bonded composite repair patch. The electromechanical impedance and acoustic stress wave methods are used to track the evolution of disbond growth. Both techniques were found to provide a reliable and robust basis for the detection of disbond growth.

This paper addresses the use of a transfer function technique for crack detection in a metallic plate. It will be shown that an array of sensor/actuator is best suited for this application. This paper will also present a simple technique of analyzing the data obtained from this array of sensor/actuator to approximately locate the position of the crack. Some experimental results will be presented.

Initially an automated sensitivity-based shape optimisation procedure is developed for the optimal design of free-form bonded repairs and lap joints, with the aim of achieving reduced adhesive stresses. The finite element approach is demonstrated through application to both a double-sided lap joint configuration, as well as for the reinforcement to an F/A-18 bulkhead. Here both the shapes of the adhesive layer and the outer adherend are allowed to vary. Significant improvements over conventional designs are obtained, as assessed by the reduction in peak adhesive stresses. This is followed by analysis to investigate stress magnification effects due to in-plane load attraction arising from bonded doublers of various shapes in remotely loaded plates. The optimality of the candidate shapes is then discussed.

A novel shape optimization scheme is presented for minimizing the adhesive stresses in bonded patch to curved thin-walled structures. Sequential linear programming is employed to search the optimum shape of the bonded patch. In each iteration, finite element method is employed to conduct stress analysis and sensitivity analysis using a novel adhesive element. An automated mesh generation algorithm is developed to adapt the change of geometry of the patch. Numerical examples show that the peak adhesive stresses can be reduced remarkably through optimizing the shape of bonded patch.

Adhesive bonding has been used to join or repair metallic and/or composite structural components to achieve or restore their designed structural stiffness and strengths. It has been demonstrated in numerous programs over many years that the critical interlaminar adhesive stresses govern the lifetime performance and durability of adhesively bonded joints and repairs. Current design tools for composite bonded joints and repairs typically use laminate level analyzes in which the properties of the laminate are homogenized and the residual ply level stresses are neglected. This study uses the BSAM spline variational three dimensional ply level stress analysis method to model a double-lap shear specimen. Moiré interferometry was used to measure the surface strains on the edge of the specimen in the bondline end region. A comparison of the predicted and experimentally measured strains shows excellent agreement.

The spline approximation approach for three-dimensional stress analysis in laminates containing open holes has been extended to account for matrix cracking in each ply. Crack surface where the displacement jump takes place is defined by using the domain Heaviside function approximated with higher order polynomial B-splines. It is shown that the spline approximation of the Heaviside function maintains the integral properties of the step function and that the gradient of the approximation maintains the integral properties of Dirac's delta function for any order of approximation. An advantage of the proposed method is that its implementation only involves integration of the products of original shape functions and their derivatives and does not require modification of the integration domains. Uniaxial tension of a unidirectional composite with open hole is considered. Fiber direction stress relaxation due to longitudinal splitting is successfully modeled by the method proposed. Observed by using incremental x-ray technique split origination and extension loads correlate well with predictions.

Residual strain predictions for an adhesive layer between composite adherends are performed using piece-wise linear elasticity. The analysis was performed using an in-house stress analysis program based on a higher order displacement spline approximation method. The residual strain measurements were conducted by using the moiré interferometry technique. A diffraction grading was installed at room temperature through the entire thickness of a polished edge of the fully cured adhesive joint. The joint was cut through the thickness along a vertical cross-section in the middle of the diffraction grating area. A full field deformation pattern was obtained in the grating area by analyzing the recorded fringe patterns. The deformation field induced by the cut is caused by the residual stress redistribution and can be predicted by the piece-wise linear elasticity analysis outlined above. A good agreement between the analysis and the experiment was observed. It is noted that the temperature dependent stress relaxation in the AF-191 adhesive is critically important for accurate residual stress prediction. Linear analysis based on room temperature thermo-mechanical properties leads to overestimation of the residual stress values.

As two dissimilar materials are bonded or co-cured at elevated temperatures, residual stresses result upon cooling the layered material system to room temperature. The residual stresses result from the thermal expansion mismatch between the materials. Current closed form solutions and experimental methods used to determine the stresses yield average stresses in the center of the material, away from edges or discontinuities. It is well known that the free edges of composite laminates possess interlaminar shearing stresses experienced during applied mechanical loading or from processing of the material. These shearing stresses are significant and may lead to failure. Until now, no experimental technique was available to assess the stress-state on the free edge. This paper describes the development of a novel experimental technique using moiré interferometry that is able to determine the residual stresses resulting from elevated temperature processing at the free edges of layered materials. The novel experimental method, referred to as Material Removal for Free Edge Evaluation (MRFEE), combines the versatility of moiré interferometry with a destructive material removal procedure to assess the free edge stress distribution.

Assemblies of topologically interlocked tetrahedron-shaped elements form layers in which each individual block is held in place by neighbouring blocks. Under lateral constraint, such assemblies keep their integrity without any binder phase or adhesive. Failure of a single block cannot cause failure of the assembly, as with a single missing or damaged block interlocking within the assembly is still retained. Weak adhesion between the blocks (or total absence of adhesion) arrests propagating cracks, induced e.g. by impact, and prevents them from spreading into neighbouring blocks. For block fracturing occurring at random and in an uncorrelated way, the total assembly disintegration will require fracturing of as many as about 59% of all blocks. This suggests the suitability of the proposed layers of interlocking elements as shields against impacts. The large fraction of 'vacancies' the system can tolerate makes it possible to fill them with blocks of different material that does not have to possess good mechanical properties and can be used to fulfil other functional requirements (heat insulation, sound absorption, etc.).

A plate containing a circular hole, and reinforced by a patch may as a first approximation, be analysed by treating the reinforced structure as a plate containing an inclusion. This leads to a state of uniform stress in the inclusion. In this paper, an exact solution is obtained using a layered-plate theory where a general remote biaxial loading is split into a combination of hydrostatic tension and pure-shear. The focus is on the pure-shear component where it is found that the stress in the reinforcement over the hole is not uniform, contrary to the expectation from the two-dimensional inclusion analogy.

An elasto-plastic analysis of the three-dimensional stress-state at blunt notch root is presented, with an emphasis on the influence of plate thickness on the stress distributions. By assuming that the through-the-thickness extensional strain is uniform in the thickness direction, explicit solutions of the three-dimensional stresses are obtained for blunt V-notch with a circular tip using the generalised plane-strain theory. Making use of the elastic solutions and the deformation theory of plasticity, an analytical method is developed to predict the elastic-plastic stress distribution ahead of a notch tip in a finite thickness plate.

This paper presents exact solutions for buckling of multi-span rectangular plates. The plate is assumed to be simply supported on two opposite edges and the other two edges may take any combination of free, simply supported or clamped conditions. The Levy solution procedure in connection with a domain decomposition technique is employed to develop an analytical approach for the buckling analysis of multi-span plates. Buckling factors, most of which are first-known exact solutions, are given in tabular form for several multi-span plates subjected to uni-axial in-plane load in the x or y directions and bi-axial in-plane load. The exact buckling solutions presented in this paper are of benchmark values for such plates.

This paper is concerned with the elastic buckling of triangular plates having translational and rotational elastic edge supports. The energy functional of a general triangular plate with elastic edge supports is derived and the p-Ritz method is utilised to obtain the governing eigenvalue equation for the buckling problem. First-known buckling factors are presented for several selected isosceles triangular plates of various edge support conditions and subjected to isotropic in-plane compressive load.

The phenomenon of hull slamming results when the bow of a ship emerges from the water and then submerges again with a vertical velocity greater than a certain threshold velocity. If the bow geometry of the ship is such that the angle between the bottom plates and the water (dead rise angle) is small, this action can produce large forces for short time durations. Traditionally, a statistical approach has been used whereby the maximum slamming load is applied as a static force. New numerical techniques are available which model the ship and waves as interacting dynamic components in the scenario. Multi-materials, Eulerian methods when coupled with Lagrangian procedures can be employed to simulate this fluid structure interaction. A generic example of a ship structure impacting a wave profile will be used as an example to demonstrate the application of these non-linear dynamic finite element methods.

The objective of this paper is to propose a general theory for multilayered shells with actuation strain. The constituent layers in the shell are in perfect bond, and the actuations strain is constant within every layer. A layer-wise representation of the displacement field is assumed, and the compatibility condition of traction at layer interfaces and on the major surfaces is imposed to reduce the number of unknowns. Hamilton's principle is used to derive the governing equations that involve only five unknowns, as in the first-order shear deformation theory. Dynamic effect and the Karman-type nonlinearty are incorporated. Application of the present theory in the analysis of some smart structures integrating actuation layers is discussed.

In this paper, effects of vertical and round beads used for shear webs and ribs in the composite structures are studied. Using these methods, we can achieve more lightness and lower manufacturing cost. In this investigation, instability (buckling) of wing nose and main ribs made of woven E-glass material and/or P.V.C foam core with and without beads are studied using finite-element analysis and experimental methods. The results show that we can easily improve shear webs stability using vertical male beads and/or round beads instead of sandwiched webs. It may cause weight reduction of about 50%, manufacturing time reduction of about 50% and increasing the buckling load. Further studies and tests on optimum ply angles on buckling and post buckling of these new structures show that by introducing 0° and 90° plies in the core region of the laminates, higher buckling load is obtained, but substantially post buckling stiffness is increased.

A semi-analytical approach called the finite strip method (FSM) is used in this paper to investigate the effects of interaction of in-plane compression and shear loading on the buckling load capacity of blade-stiffened composite panels. The panel is loaded under pure compression, pure shear and combined loading. The interaction curves are plotted for different stiffener aspect ratios. Also, the buckling mode shapes for a few selected points on the interaction curves are plotted. Based on the obtained results certain conclusions are made.

This paper identifies different shear band orientations for different types of steel tubes and sheet steels. Strain gauge rosettes were used to measure the maximum shear stress orientation for very high strength (VHS) tubes. It was found for VHS tubes that a final 62.5° (rather than 45°) shear band developed during the plastic deformation stage. This paper explored the reasons behind the different orientations, including anisotropy of steels, rotation of steel crystals during plastic deformation, material properties, manufacturing process and geometric effects.

Radial prestress in the grout is generated by adding an agent to the Portland cement so that it expands when setting. A normal constituent of cement, a complex Calcium Sulpho-Aluminate (CSA) is used. Prestress is a function of the expansion of grout which may depend on many factors such as the percentage of CSA in the mix, water cement ratio, time, surface condition of sleeve tubes, grout thickness, grouted length and ratio of tube diameter to thickness. There are insufficient data to establish a formula to predict prestress over the full range of connection parameters. The existing data only covers limited parameter ranges. A series of tests was designed to include a wide range of percentages of CSA, water cement ratio and grout thickness. Strain gauge readings were used to obtain measured prestress. A long-term data acquisition system was used to monitor the readings. The newly obtained results were used to establish a formula for confidently predicting the prestress in grouted sleeve connections and the existing data were used to verify this equation. The prestress formula will be used in predicting the capacity of the grouted sleeve connections in tension or compression.

A finite-element-based gradient-less shape optimisation procedure has been used to develop free-form optimal rework shapes for a number of critical high stress concentrating features in the F-111 wing pivot fitting structure. A gradient-less method was used here, in preference to the design sensitivity method that is more commonly found in commercial finite element software, due to its simpler approach and compatibility with higher order p formulation elements. The use of higher order elements resulted in a very accurate and timely efficient stress solution as compared to standard h formulation elements. An added benefit of the relatively large p formulation elements was that significant boundary movements could be made with minimal element distortion, thus reducing the number of re-meshes required during the optimisation procedure. The resulting optimal shapes provided a predicted 30% reduction in peak elastic stresses as compared to traditional shapes consisting of circular arcs. These predictions have been confirmed by a recent full-scale static wing test. It is anticipated that these stress reductions will significantly reduce the risk of fatigue cracking, thus allowing the airframe to reach the planned withdrawal date of 2020. In addition, a lower inspection frequency is expected, which will provide large maintenance cost savings and increased aircraft availability. Fleet-wide implementation of the optimal rework program is scheduled to commence in 2001, as part of the F-111 Australian Sole Operator Program.

This paper looks at the design of the bicycle using the Evolutionary Structural Optimisation (ESO) method. It gives a brief overview of the history of the bicycle and a description of its anatomy. It then goes on to carefully determine equations that predict all of the riding loads experienced by the bicycle. These are applied onto a design domain and by using the ESO method various examples are produced and analysed.

The paper first summarises the linear theory for determining displacements and internal forces developing from earthquake motion. The analysis can be a transient dynamic analysis where a ground acceleration history is used for calculating displacements and corresponding internal forces. Alternatively, the analysis can be a spectral analysis in which an equivalent external static force vector is calculated which can be used as loading in a linear static analysis. The two types of analysis constitute the "dynamic" analyses specified in the Australian earthquake standard AS1170.4. The paper describes the implementation of dynamic earthquake analyses in the general purpose program PRFSA and includes an example of a spectral analysis of a steel frame.

This paper presents and discusses the applications of neural networks in optimum design of concrete structures. It aims to introduce some of the neural networks applications, especially structural design. Backpropagation networks are chosen for the proposed network, which is written in the programming package MATLAB. The overall results are compared and observed for the performance of the networks. Based on the application it is found that neural networks are comparatively effective for a number of reasons, which include the amount of computer memory consumed by neural networks is less than that consumed by conventional methods and their ease of use and implementation. Neural networks provide both the users and the developers more flexibility to cope with different kinds of problems. One of the significant premises to this reason is that the modification of neural networks is much easier than the modification of most conventional methods. When many different problems are to be solved, the equations used by certain networks do not change. This results in the reduction of implementation time. Consequently, the cost of their implementation is relatively low.

This paper presents a formulation of Genetic Algorithm based optimization for trusses. The minimum weight of trusses with geometrically non-linear behaviour is sought. Constraints are imposed on member stresses and nodal displacements when the member cross-sectional areas are available from a discrete set of a given catalogue. The GA is based on a roulette-wheel reproduction scheme, a one-point crossover, and a standard mutation scheme. An elitist strategy is also used that passes the best designs of a generation to the next generation. GAs are well suited for unconstrained optimization problems, a transformation based on the violation of normalized constraints is used to transform the constrained problem into an unconstrained one. An incremental load approach with a Newton-Raphson type of iteration is used in the geometrically non-linear analysis procedure and in developing the optimization algorithm. Numerical results are presented that show the efficiency of the optimization procedure.

With the aid of molecular dynamics analysis, this study uses a three-asperity model to investigate the effects of the relative orientation and position of the asperities on the nano-wear mechanism of silicon. It was found that when the first asperity has created a damaged layer, the material will deform differently under subsequent sliding. A thin amorphous layer always remains and there is also an absence of dislocations when the depth of asperity penetration is small. On the other hand, the forces experienced by the asperities differ and are dependent on their relative positions. The results suggest that microstructural changes in silicon due to the initial sliding affect the subsequent deformation over a damaged region.

Elastohydrodynamic Lubrication (EHL) theory is frequently used to determine the most suitable lubricant for gear drives. Although it is known that the visco-elastic properties of the lubricant will affect the results of computation, a Newtonian model is chosen for simplicity. In particular, the lubricants used for open type gearing are of a sprayable, adherent type and are usually significantly non-Newtonian. Attempts to correlate the performance of open gears in the field with EHL theory based on Newtonian behaviour have been unsuccessful. This is due mainly to non-Newtonian effects. This paper reports on the Theological properties of some open gear lubricants and highlights the vast differences between the products which are supplied to perform the same task. The complexities which arise when using a grease, or grease-like, lubricant are gone into in some detail. It is concluded that, for thick film lubrication, the Maxwell relaxation time may be more important and the base oil viscosity less important than hitherto assigned.

Environmental concerns have resulted in a need for cleaner fuels. Current requirement limits the sulphur content of diesel to 500 parts per million. It is expected that the limit will be set at a low 50 parts per million by 2006, although it is expected that the industry will move to the 'ultra low sulphur' diesel before then. Diesel with low sulphur levels produces less black smoke emissions and less of the smaller particulate matter, which has adverse health effects. Removing sulphur from diesel results in the reduction of some aromatics in diesel during the distillation. This has an adverse effect on the lubricating properties of diesel fuel. The shortened life of engine components such as the fuel injection pump and injectors has been attributed to the lack of lubricity of diesel fuel. Experiments are conducted using a High-Frequency reciprocating rig built as per ASTM D-6079-99 [1] on 5 samples of diesel with sulphur content in the range of 245 to 480 parts per million. Experiments are also conducted when a readily biodegradable lubricity additive is used in each sample. The wear scar area and the coefficient of friction are measured in each case. The effect of low sulphur levels and additive on wear and friction coefficient are studied and compared. Typical results obtained are presented.

In most of the manufacturing processes for continuous fibre reinforced plastics, the dry fibres must be wetted by a liquid matrix, such as thermosetting resins or thermoplastic polymer melt or solution. After the solidification process, quality requirements demand a solid composite part without voids. Economical constraints are making this task to a compromise between a fast process and a good quality of the part. In order to reduce experimental efforts, computational flow simulations are very common in optimisation of the manufacturing processes. For the mathematical compilation of the problem, Darcy's law is used most of the times. The method requires a factor, the permeability. This fibre bed related constant must often be determined experimentally or is roughly estimated. In this work, a new approach for determining permeabilities, taking into consideration of the stochastic characteristics of the fibre bed, is presented.

A novel method for generating compressive residual stresses on surfaces of metallic alloys is laser shock peening (LSP). The LSP process uses a high intensity, short duration laser pulse to create shock waves in the materials. LSP was performed on an aluminum alloy with a Nd-glass laser beam of 8 mm in diameter, operating with a 50 J per pulse and 25 ns in duration. Shock waves were generated by volume expansion of plasma when the material was laser-irradiated. The specimen was coated by the energy-absorbing black paint, and water as the plasma-confining layer was used to obtain the maximum pressure without melting the specimen surface. The effect of laser shock peening on the residual stresses of the aluminum alloy was studied and simulated using the finite element method. The finite element codes (the ABAQUS standard and ABAQUS explicit) were used to simulate the two-sided peening mechanisms on the specimens.

There are several industry-attractive aspects of flux cored arc welding (FCAW) process that needs further investigation. Also, the welding parameters that can provide the required microstructure, hardness and minimum residual stresses in thick walled weldments need to be established to validate the application of flux cored arc welding process.

In the present work, the effect of welding parameters and type of shielding gas on the weld bead geometry, microstructure, hardness, spatter rates and depth of penetration have been studied and reported. The results shows that the spatter rates increase as the CO₂ content of Ar+CO₂ shielding gas mixtures increases. The hardness of all the weld metals is affected by the amount of proeutectoid ferrite and the size of acicular bainitic structure of the columnar grains of deposited metal. On the other hand, the changes in hardness values at the heat-affected zone (HAZ) were found to be a function in the changes of the heat input. In this paper, values of welding parameter were obtained to give best quality of welding using FCAW process.

Most modern cars have steering system utilising Arthur Bishop's rack and pinion mechanism. When it wears out replacement is costly. Attempts to re-manufacture them were unsuccessful because teeth have variable module and height. As a result interference takes place on re-manufactured mechanisms when the rack travels from mid-position. This may lock the steering system with catastrophic consequences. To investigate the meshing of virgin and remanufactured rack and pinion "Mechanica" software was used ("Motion" module). The clash points were detected and recommendations for profile modification were developed to prevent interference.

In conveyors utilising the principle of semi-self-flowing transportation parts are put on the surface moving in the lateral direction by means of vibration or on two rollers rotating in opposite direction. Advantages of the latter one are that parts can be supplied on demand, and very small angle of inclination is required (less than 3°). Formulas found in scientific literature for the velocity of transportation as a function of the roller velocity of rotation give large discrepancy with experimental results. Extensive experimental modelling allowed to establish the dependence of the frictional coefficient on the shape of parts and the contact angle. A new formula was derived which is in good agreement with experimental data.

A piece of cloth can be modelled physically by using a mass spring network. The realism and accuracy of the model are largely dependent on the network characteristic, which are defined by the network topology and spring stiffness. This paper presents the derivation of these two quantities based on the strain energy of the cloth. The fabric properties are incorporated into the network characteristic. As a result, distinguishable fabric can be simulated realistically. Draping examples are raised to compare with the real situation.

The paper describes methods used to determine the rigid body mode shapes and their corresponding natural frequencies and damping values of an inter-urban rail carriage on its own suspension. For bounce, pitch and roll, the method used to induce the motion was to pull specified wheels (in different combinations) up onto chocks of thickness 15 mm, and then let them fall back onto the rail simultaneously. For yaw, the carriage was impacted laterally at one end by a tractor, cushioned by rubber tyres. To analyse the response signals, the effective step responses were treated in two ways. Natural frequencies and (operating) mode shapes were determined approximately by ODS (Operating Deflection Shape) measurement based on measuring relative motions at a specified frequency as close as possible to the natural frequency for that mode. The method used to give the best estimates of mode shapes, frequencies and damping was to treat the effective step responses as though they were impulse responses, since these would still have the same poles. This allowed determination of the global parameters by curve fitting all four response measurements at the same time, and then fitting the mode shapes based on these global parameters. An exponential window had to be used to extract the best part of the response, but the resulting added damping could be subtracted from the final result.

This paper examines the response reduction capabilities of a passive fluid damper device, known as the Liquid Column Vibration Absorber (LCVA), when attached to a five storey benchmark building structure requiring control. The five storey benchmark building is an experimental model that can be excited at its base level to simulate ground motion. In previous studies the LCVA has been shown to be effective for wind vibration control. In this study the probable structural response reductions that can be achieved due to four benchmark earthquakes is examined. A further investigation is made to the use of multiple LCVA's (MLCVA) for earthquake response suppression. The issue of robustness is also addressed by comparing the response reduction capabilities of the LCVA and MLCVA configuration under earthquake loading. The robustness becomes a primary issue when the LCVA may become largely mistuned in the event of loss of structural stiffness during an earthquake.

This paper presents the second stage of applying an active control scheme to the benchmark model of five stories. The first stage reported the use of an Active Tuned Mass Damper (ATMD) system on the benchmark model, where the control action is achieved by a Fuzzy Logic Controller (FLC). In this stage, the active control scheme is applied to the modified benchmark model equipped with an Active Mass Driver (AMD) system, where the control action is again achieved by using a Fuzzy controller. In addition, the experimental tests on the AMD system is reported. The building model under consideration is the 5-storey, 3.6 meters tall, steel frame designed and manufactured at the University of Technology, Sydney. In order to test, experimentally, the five-storey benchmark model, an Active Mass Driver (AMD) system was designed and manufactured at the University of Technology Sydney and the performance of the AMD hydraulic system was verified analytically. In this stage, the experimental tests are conducted on the AMD system to verify the analytical predictions before using the system on the benchmark model. Performance of the controller is validated through computer simulations. The advantage of the Fuzzy controller is its inherent robustness and ability to handle any non-linear behaviour of the structure. The results of the simulation show good performance of the Fuzzy controller for all four earthquakes used in the simulation.

This paper presents vibration control of a five-storey model with semi-active stiffness damper subjected to earthquake loads. Semi-active stiffness damper consists of a special type of hydraulic damper connected to the bracing frame and installed in a selected storey unit. Various control laws are applied to demonstrate the system performance. They are resetting control, switching control and linear quadratic regulator (LQR). Three cases of resetting control are developed by selecting control parameters to explore control performance. Moreover, the damper effective stiffness is varied to determine the optimum performance. The results show that semi-active stiffness damper effectively controls the vibration of a five storey model. By selecting an appropriate damper stiffness value, each control law can perform well. Resetting control can achieve maximum displacement reductions of up to 83%, while it is 73% for switching control and 74% for LQR.

Base isolation has been used in many classes of civil engineering structures to reduce the damage from earthquake attacks. The main idea behind this innovative technique is to shift the fundamental natural frequency of the structure to a value much lower than the dominant frequency of the earthquake. In this work, dynamic characteristics of a five-storey benchmark model isolated with laminated rubber bearings (LRB) and Lead core rubber bearings (LCRB) under Kobe and El-Centro ground motions were examined using a shaking table. The earthquake resistant performance of LRB and LCRB isolators was evaluated. The first natural frequency of the model was reduced from 4.7Hz for the unisolated model to 2.9 and 1.7 Hz for the LRB and LCRB isolated models, respectively. It was observed that maximum floor accelerations were significantly reduced with the addition of an LRB or LCRB isolation system regardless of the ground motion input. However, the LRB was identified to be more effective than LCRB in reducing accelerations and model relative displacements, and therefore, provided a better protection of the superstructure and its contents. On the other hand, the LCRB produced smaller base and floor displacements, and hence can ensure structural stability better than the LRB. This property of LCRB stems from high damping and stiffness of the lead core.

Among the varieties of control algorithms, Variable Structure Control (VSC) or Sliding Mode Control (SMC) is known for its most distinguished feature and ability to result in ultra-robust control systems with respect to parametric uncertainties of structures. In this paper, control methods based on Variable Structure Control (VSC) are presented and applied to a seismically excited civil engineering structure equipped with shape memory alloy actuators. Design of the sliding modes including the eventual sliding mode, was accomplished by using LQR methods. In developing control algorithms, both Lyapunov direct method and reaching law methods were used, which result in four control laws. A comparison of these control laws assisted in further understanding of Sliding Mode Control theory. Simulations were conducted to evaluate the level of vibration reduction using the presented control methods.

A model of the free and forced vibrations in a periodically supported infinite rail is developed. Periodic Structure Theory is used to limit the extent of the model to a single rail segment and its support. The model incorporates degrees of freedom internal to the rail segment, allowing the rail support to be represented in detail. The Finite Element method is used to model the stiffness and inertial properties of the rail, enabling the method to be applied to beams of arbitrary cross-section. The dispersion characteristics and driving point receptance for an infinite equal-angle beam are calculated and compare favorably with properties derived using a simple analytical model. The newly developed model is applied to a prototype rail geometry and the results discussed.

A method is developed to obtain the natural frequencies of a beam with any number of changes in cross-section and applied to an n-step beam of constant depth consisting of (n+1) equal portions with the breadth changes at the steps in arithmetic progression. The procedure enabled any number of steps to be tackled without encountering any computational difficulties. Table of the first three frequencies of a clamped-free and a free-clamped stepped beam with 1, 2, 3, 4, 5, 10, 50, 100, 200 steps are tabulated. It is shown that as the number of steps is increased, the frequencies asymptotically approach the frequencies of a 'truncated' beam of constant depth and linearly varying breadth.

This study aims to develop a comprehensive mathematical model and a simulation system of an automatic transmission to investigate the transient characteristics during gear changes. In particular, the model is applied to a 4-speed automatic transmission manufactured by BTR Automotive to study the 2-3 shift process. A brief description of the BTR automatic transmission and the 2-3 shift mechanism are presented. A mode description method is used for describing the shift process. The modular structure of the simulation system, which includes engine module, torque converter module, hydraulic system module, geartrain module and modules of clutches and bands, is developed, and the governing equations of motion of the integrated powertrain system are derived. The simulation results are presented and the major parameters affecting the quality of the shift process are discussed.

This paper describes the capability of transient analysis software to simulate rolling element bearing defects in a simple rigid rotor, flexible housing configuration. Numerical examples using a deep groove ball bearing are used to illustrate via rotor displacement and pedestal acceleration amplitude spectra and wave forms the effect of an inner or an outer race defect under gravity and/or unbalance loading. The different effects of inner and outer race defects on the vibration behaviour of the bearing are clearly illustrated.

Shape reworking can be a highly effective repair methodology to extend the life of metallic airframe components suffering fatigue cracking at stress concentrations. In recent years a gradientless shape optimisation technique has been successfully used to determine precise free-form optimal rework shapes, which minimise the peak stresses. In this paper some recent generic and practical applications are first described, such as the optimisation of generic tension fillets, as well as an optimal rework design for the PC9 wing skin. Subsequently the issue of sensitivity of optimal shapes to potential manufacturing errors is discussed, using as an example the case of perturbations to an optimal rework hole in the F-111 wing pivot fitting. The results obtained indicate the suitability of the present approach to yield high quality optimal solutions, noting that typically, optimal shapes are more sensitive to manufacturing error than standard shapes, and hence care must be taken.

In-service piping is often exposed to through-wall temperature gradients. The effects of these on the thermal stresses and creep response are examined using the Theta projection creep algorithm within a finite element code. It was found that the stress and temperature dependence of the creep response may lead to complex stress evolution.

The following Sections are included:

• AUTHOR INDEX