Search

Powell–Sabin B-splines, which are based on triangles, are employed in the framework of the extended finite element method (XFEM) for fracture analysis. This avoids the necessity of remeshing in discrete fracture models and increases the solution accuracy around the crack tip. Powell–Sabin B-splines are C 1 -continuous throughout the whole domain. The stresses around crack tips are captured more accurately than when using elements with a standard Lagrangian interpolation. Although Powell–Sabin B-splines do not hold the Kronecker-delta property, the Heaviside function and the tip enrichment function are confined to the cracked elements only, similar to the traditional XFEM but different from the extended isogeometric method. In addition, Powell–Sabin B-splines still hold C 1 -continuous throughout cracked elements. There is no need to lower the continuity at element boundaries, to confine basis function support in cracked elements. Shifting is used to ensure compatibility with the surrounding discretization. The sub-triangle technique is employed for the numerical integration over crack elements. The versatility and accuracy of the approach to simulate crack problems are assessed in case studies, featuring mode-I and mixed-mode crack problems. • Extended Powell–Sabin finite element method is developed for fracture analysis. • The extended scheme confines the effect of the crack in cracked elements only. • Shifting is proposed and elaborated to enforce compatibility for higher-order bases. • Numerical integration scheme is proposed for the integration in the cracked elements. [ABSTRACT FROM AUTHOR]

This paper presents a highly efficient small-scale, detailed finite-element modelling method for flexible risers which can be effectively implemented in a fully-nested (FE 2 ) multiscale analysis based on computational homogenisation. By exploiting cyclic symmetry and applying periodic boundary conditions, only a small fraction of a flexible pipe is used for a detailed nonlinear finite-element analysis at the small scale. In this model, using three-dimensional elements, all layer components are individually modelled and a surface-to-surface frictional contact model is used to simulate their interaction. The approach is applied on a 5-layered pipe made of inner, outer and intermediate polymer layers and two intermediate armour layers, each made of 40 steel tendons. The capability of the method in capturing the detailed nonlinear effects and the great advantage in terms of significant CPU time saving are demonstrated by comparing the results obtained on elements of pipe of different lengths, equal to one pitch length L p as well as L p / 5 , L p / 20 and L p / 40 . [ABSTRACT FROM AUTHOR]

Abstract: An updated Lagrangian finite element formulation of a three-dimensional annular section beam element is presented for large displacement and large rotation dynamic analyses of flexible riser structures. In this formulation a new linearization method is used to avoid inaccuracies normally associated with other linearization schemes. The effects of buoyancy force as well as steady state current loading are considered in the finite element solution for riser structures response. The formulation has been implemented in a nonlinear finite element code and the results are compared with those obtained from other schemes reported in the literature. [Copyright &y& Elsevier]

Abstract: This article proposes a predictive method for identifying the range of sea-states considered safe for the installation of offshore structures. A finite element dynamic analysis of the system for various sea-states characterized by significant wave heights and mean zero-up-crossing wave periods and modeled as a combination of several wave components has been performed. Using this procedure a table of safe and unsafe sea-states is generated. The significant wave height (H s ) and mean zero-up-crossing wave period (T z ) of a future sea-state in a location in the north east Pacific were predicted from the distributions whose parameters were estimated using the artificial neural networks (ANNs) trained for this purpose. The location of US National Oceanographic Data Center (NODC) Buoy 46005 is used in this study. The H s and T z of some future sea-states were predicted from their corresponding conditional 7-parameter distribution given some information including a number of previously measured H s ’s and T z ’s. This gives a predicted sea-state for a specific time in future. The parameters of the distributions have been estimated from the outputs of two different 7-network sets of trained ANNs. A pile-driving operation is used as a case study in which the pile configuration, including the non-linear foundation and the gap between the pile and the pile sleeve shims, has been modeled by the finite elements method and the range of sea-states suitable for safe pile-driving operation was identified. [Copyright &y& Elsevier]

The derivation of the long-term statistical distribution of significant wave heights (Hss) is discussed in this paper. The distribution parameters are estimated using artificial neural networks (ANNs) trained with the help of a simulated annealing algorithm and operated in an autoregressive mode. The ANNs were utilized in estimating the parameters of a conditional probability distribution related to a desired Hs given its preceding Hss, approximated by a proposed distribution called the hepta-parameter spline. The performance function during training was based on the likelihood function of the statistical method of maximum likelihood estimation (MLE). Given the observed dataset, the most probable weights and biases of the neural networks were determined in such a way that the performance function was optimized. The distribution could be used in the simulation and forecasting of Hss. This paper also presents an extreme wave analysis using the simulated Hss. The extreme analysis conducted in this study using the maxima method offers an alternative approach, avoiding the unrealistic hypothesis that annual Hss are identically distributed, as is conventionally assumed when using the Fisher–Tippet theorem. [ABSTRACT FROM AUTHOR]

This article presents an inverse formulation of the eigenvalue problem for computing the required changes in laminated composite plates in order to achieve desired dynamic characteristics in the structure. The stiffness and mass matrices of the plated structure are first derived using the finite element formulation based on the first-order shear deformation theory (FSDT) for laminated composite plates with arbitrary angle-ply stacking sequence. Based on this formulation and using the first- and second-order Taylor expansion both the direct and inverse eigenvalue problems are formulated to find the changes in eigenvalues due to an arbitrary change in physical or geometrical properties of the structure. An initial sensitivity analysis is conducted to identify the layer in which modification of design variables have the most influence on the structures dynamic characteristics. The design variables in this context are defined as the fiber angles in each layer and the layer thickness. The proposed algorithm is applied to several case studies to demonstrate the application and the accuracy of the proposed formulation in solving the direct and the inverse eigenvalue problems with one and two design variables. Dynamic behavior modification of a plate is performed using the inverse approach, and the required modified stacking sequence is obtained to shift the natural frequencies to desired values. [ABSTRACT FROM AUTHOR]

Abstract: The ability to model acoustic emission (AE) plays an important role in advancing the reliability of AE source characterisation. In this paper, an efficient numerical approach is proposed for modelling AE waves in isotropic solids. The approach is based on evaluating the reciprocal band-limited Green''s functions using the finite element (FE) method. In the first section, known analytical solutions of the Green''s function for an elastic isotropic infinite plate subjected to point monopole surface loading are used to validate the approach. Then, a study investigating the effects of the spatial resolution of the FE model on the accuracy of the numerical solutions is presented. Furthermore, comparisons between numerical calculations and experimental measurements are presented for a glass plate subjected to two known AE sources (pencil lead break and ball impact). Finally, the reciprocal relation between the source and the receiver is confirmed using numerical simulations of a plane stress model of an elastic isotropic plate. [Copyright &y& Elsevier]

A cylindrical shell element is developed for the calculation of natural frequencies of cylindrical panels as well as complete cylinders. The element is based on assumed strains rather than displacement and possesses the only five essential degrees of freedom. The displacement fields of the element satisfy the exact requirement of rigid-body modes.Convergence tests of the mesh discretisation are first carried out to establish the rate of convergence of the element. These have been applied to simply supported as well as clamped cylindrical panels. The tests show that good rate of convergence can be obtained when only a small number of elements are used. The element is then further applied to the analysis of complete cylinders and a comparison is made with available solutions. [Copyright &y& Elsevier]

This paper extends the first order formulations presented in Part I to second order methods for relocation of structural natural frequencies from their initial design values to new modified frequencies. The method is based on an inverse formulation and solution algorithm of the eigenvalue problem. Using the second order Taylor''s expansion series, the required parameter variation to achieve a desired natural frequency shift for the structure is computed through second order differential or binomial equations. The proposed technique can also incorporate the design constraints or objective functions in the system equations. The formulations are quite generic and applicable to all finite element structures. The accuracy of the proposed methods is tested by conducting several case studies, the results of which demonstrate the validity of the technique for a wide range of practical problems. [Copyright &y& Elsevier]

This paper proposes an inverse strategy for relocation of structural natural frequencies using first order formulation and solution algorithm. In the proposed method a sensitivity analysis of the systems eigenvalues with respect to material or geometrical parameters of the structure is conducted initially. The required parameter variation to achieve a desired frequency shift for the structure is then computed. The proposed technique incorporates the design constraints or objective functions in the system equations in such a way that a square system of equations is always preserved. The formulations are general and applicable to all finite element structures because the sensitivity analysis is based on the stiffness and mass matrices regardless of the type of elements used. An algorithm suitable for implementation of the technique for practical purposes is developed. The accuracies of the proposed methods are tested conducting several case studies and the results are validated against exact solutions. [Copyright &y& Elsevier]

This paper investigates the effect of cut-outs on the dynamic behaviour of cylindrical panels, using a newly developed strain based finite element. The developed element is based on assumed strains and has only five necessary degrees of freedom at each corner node. The displacement fields of the element satisfy the exact requirement of rigid body displacements. The efficiency of the element is tested by applying it to the calculation of natural frequencies of shells. Investigations are first carried out to test the convergence of the element and to establish the mesh size to be used. The element is further applied to analyse cylindrical panels with cut-outs. Simply supported as well as clamped panels on their edges were considered and the effects of the size and location of the openings on the natural frequencies are investigated. The subspace iteration technique which is shown to have an economising effect, is used to obtain the natural frequencies and the associated modes of vibration. [Copyright &y& Elsevier]

This paper presents a formulation in the form of an inverse eigen value problem for modification of vibration behavior of structures. The proposed method which is based on the second order approximation in Taylor expansion is expressed in terms of variables relating to stiffness or mass matrix parameters in a finite element formulation. An initial sensitivity analysis identifies the regions within the structure where the modifications would yield the required changes in the structures dynamic characteristics. An algorithm is developed which allows efficient modification of structural dynamics characteristics without iterations. These modifications are conducted locally so that only elemental stiffness and matrices are affected. The algorithm is applied to four case studies and it is found that large modification of natural frequencies of up to 10% can be realized with an induced error of less than 5% for truss structures, and less than 3% for plane problems. [Copyright &y& Elsevier]

This paper presents a cylindrical strain based shallow shell finite element which is developed for linear and geometrically non-linear analysis of cylindrical shells. The developed element is rectangular inplan and has the only five essential degrees of freedom at each corner node. The displacement fields of the element satisfy the exact requirement of rigid body modes. The efficiency of the element is demonstrated by applying it to the linear and geometrically nonlinear analysis of shell structures. The suitability of the element is also tested for natural frequencies’ calculations of cylindrical shells. Results obtained by the present element are compared with those available in the literature. These comparisons show that efficient convergence characteristics and accurate results can be obtained by using the present element. [Copyright &y& Elsevier]

This paper presents formulations for inverse optimisation of vibration behaviour of finite element models of both truss and continuous structures. The proposed algorithms determine the required modifications on truss and continuous structures to achieve specified natural frequencies. The modification can be carried out globally or locally on the structures stiffness and matrices and the formulation can also be used to add new structural members to achieve the desired response. Numerical examples and finite element implementation of the developed method are provided to demonstrate the feasibility of the formulations. [Copyright &y& Elsevier]

This paper presents an inverse formulation of eigenvalue problem for optimising the dynamics behaviour of structures. The formulation is applied to one- and two-dimensional finite elements which are commonly used in simulation of real structures. Based on this formulation an algorithm is developed for accurate and efficient modification of structures stiffness and mass characteristics to achieve the desired natural frequencies and interpretation of results in terms of physical design parameters. The proposed algorithm is tested by conducting four case studies and the results are validated against exact solutions. [Copyright &y& Elsevier]

An efficient relationship between geometric and material properties of pin-jointed truss structures and their eigenvalues is established. The problem is formulated as an inverse eigenvalue problem. This formulation allows the determination of the required modifications on the structural members to achieve specified eigenfrequencies. In addition to the modification of the existing structural elements, the formulation also allows addition of new structural elements to obtain the desired frequencies. Using the proposed inverse method, two cases of plane as well as space truss structures are studied and the results are compared with those obtained using the conventional optimization techniques adopted by commercial finite element codes. [Copyright &y& Elsevier]

The morphology of dodecane in a nanopore at temperatures typical in exploited or depleted oil reservoirs is investigated using molecular dynamics simulation. The dodecane morphology is found to be determined by interactions between interfacial crystallization and surface wetting of the simplified oil, while "evaporation" only plays a minor role. The morphology changes from an isolated, solidified dodecane droplet to a film with orderly lamellae structures remaining within, and finally to a film containing randomly distributed dodecane molecules, as the system temperature increases. In a nanoslit under the impact of water, since water wins against oil in surface wetting on the silica surface due to electrostatic interaction induced hydrogen bonding between water and the silanol group of silica, the spreading of dodecane molecules over the silica surface is impeded by this water confinement mechanism. Meanwhile, interfacial crystallization is enhanced, leading to always an isolated dodecane "droplet," with crystallization weakening as the temperature increases. Since dodecane is immiscible to water, there is no mechanism for dodecane to escape the silica surface, and the competition of surface wetting between water and oil determines the morphology of the crystallized dodecane droplet. For the CO2–dodecane system in a nanoslit, CO2 is an efficient solvent for dodecane at all temperatures. Therefore, interfacial crystallization rapidly disappears. The competition of surface adsorption between CO2 and dodecane is secondary for all cases. The dissolution mechanism is a clear clue for the fact that CO2 is more effective than water flooding in oil recovery for a depleted oil reservoir. [ABSTRACT FROM AUTHOR]

Abstract: Ultrasonic degassing has been shown to be a reasonably efficient way of degassing a melt. Similarly, vacuum degassing has also been proven to be beneficial in producing high quality products. However, neither method is sufficiently efficient. Moreover, a combination of these techniques does not respond well to large volumes of melt. In this article, a novel approach, using ultrasonic assisted argon degassing was examined and compared with vacuum and argon degassing. The results indicate that the ultrasonic-argon degassing process is a faster method and can be used for large volumes of melt, thus overcoming the restrictions of the other two methods. [Copyright &y& Elsevier]

Experimental tests and detailed finite element analyses were carried out on a model of flexible riser to evaluate the capability of the finite element method to predict its nonlinear structural response. The riser consists of four layers, which include two cylindrical polycarbonate tubes and two steel helical layers. One helical layer represents the carcass layer in a flexible riser whilst the other represents the riser tendon armour layer. First, bending load experiments on the model are described which provide some insight regarding the fundamental behavior of flexible pipe structures. This is followed by the description of the FE models in which all layer components are individually modelled and a surface-to-surface frictional contact model is used to simulate their interaction. Finally, the FE numerical results were compared with the test data to outline the capacity of the numerical method to predict the response of flexible riser structures. [ABSTRACT FROM AUTHOR]

This paper presents an analytical formulation and a finite element analysis of the behavior of multilayer unbonded flexible risers. The finite element model accurately incorporates all the fine details of the riser that were previously considered to be important but too difficult to simulate due to the significant associated computational cost. All layers of the riser are separately modeled, and contact interaction between layers has been accounted for. The model includes geometric nonlinearity as well as frictional effects. The analysis considers the main modes of flexible riser loading, which include internal and external pressures, axial tension, torsion, and bending. Computations were performed by employing a fully explicit time integration scheme on a parallel 16-processor cluster of computers. Consistency of simulation results was demonstrated by comparison with those of the analytical model of an identical structure. The close agreement gives confidence in both approaches. [ABSTRACT FROM AUTHOR]

This paper presents a detailed finite-element analysis of unbonded flexible risers. The numerical results are compared to the analytical solutions for various load cases. In the finite-element model, all layers are modeled separately with contact interfaces between each layer. The finite-element model includes the main features of the riser geometry with very little simplifying assumptions made. The numerical model was solved using a fully explicit time-integration scheme implemented in a parallel environment on a 16-processor cluster. The very good agreement found from numerical and analytical comparisons validates the use of our numerical model to provide benchmark solutions against which further detailed investigation will be made. [ABSTRACT FROM AUTHOR]

The effect of combined electromagnetic–ultrasonic fields evaluated on grain refinement. Cavitation induced by the combination of technologies suggested as major mechanism for grain refinement. A numerical simulation performed predicting the cavitation regions in the molten AA5754 aluminum alloy. It was concluded that the cavitation phenomenon occurs near the electromagnetic and ultrasonic sources. The observed acoustic streaming induced by ultrasonic horn is more intensive than the electromagnetic one. The ultrasonic streaming could transport small bubbles into the bulk rapidly. These micro-bubbles collapsed due to acoustic vibrations where the resulting micro-jets were strong enough to break the oxide layer, wetting impurities. Exogenous particles, intermetallic and oxides may contribute to the formation of fine, uniform and equiaxed microstructure across the treated melt. Experimental results confirmed the simulation predictions. [ABSTRACT FROM AUTHOR]

Abstract: The purpose of this paper is to model the detailed effects of interactions that take place between components of un-bonded flexible risers, and to study the three-dimensional motion responses of risers when subjected to axial loads, bending moments, and internal and external pressures. A constitutive law for un-bonded flexible risers is proposed and a procedure for the identification of the related input parameters is developed using a multi-scale approach. A generalized finite element structural model based on the Euler–Bernoulli beam theory is developed in which the constitutive law is embedded. The beam theory is enhanced by the addition of suitable pressure terms to the generalized stresses to account for the internal and external pressures, and it can therefore be efficiently used for large-scale analyses. The developed nonlinear relationship between generalized stresses and strains in the beam is based on the analogy between frictional slipping between different layers of a flexible riser and frictional slipping between micro-planes of a continuum medium in non-associative elasto-plasticity. The merit of the present constitutive law lies in its ability to capture many important aspects of the structural response of risers, including the energy dissipation due to frictional slip between layers and the hysteresis response. [Copyright &y& Elsevier]

Abstract: This paper proposes a new method for performing predefined simultaneous modification of natural frequencies and buckling loads of composite cylindrical panels. The method is based on the fact that both natural frequencies and buckling loads are eigenvalues of an algebraic system of simultaneous equations. First- and second-order derivatives of these eigenvalues are calculated and the first two terms in Taylor expansion are used for developing a modification procedure that is defined as an inverse eigenvalue problem. A four-layered composite cylindrical panel with an arbitrary angle-ply stacking sequence is considered as a case study and several simultaneous modifications for natural frequencies and buckling loads are carried out. It is shown that the proposed method can perform the predefined modification with an acceptable accuracy even for large perturbations in objective functions. [Copyright &y& Elsevier]

Abstract: Grain refinement of aluminium alloys has been an established method of improving mechanical properties of aluminium alloys. A5754 wrought aluminium alloy offers a number of attractive superior properties, however, refinement of this alloy and other wrought alloys is still a matter of concern. A novel technique is introduced for the refinement of wrought and cast alloys, using A5754 alloy as a model material. This method not only provides fine and uniform microstructure throughout the products but it also offers possibilities of industrial applications. This technique eliminates the problems associated with chemical grain refinement and introduces physical refinement as a better option for achieving fine grain sizes. [Copyright &y& Elsevier]

Abstract: In this paper a new constitutive model for flexible risers is proposed and a procedure for the identification of the related input parameters is developed using a multi-scale approach. The constitutive model is formulated in the framework of an Euler–Bernoulli beam model, with the addition of suitable pressure terms to the generalized stresses to account for the internal and external pressures, and therefore can be efficiently used for large-scale analyses. The developed non-linear relationship between generalized stresses and strains in the beam is based on the analogy between frictional slipping between different layers of a flexible riser and frictional slipping between micro-planes of a continuum medium in non-associative elasto-plasticity. Hence, a linear elastic relationship is used for the initial response in which no-slip occurs; an onset-slip function is introduced to define the ‘no-slip’ domain, i.e. the set of generalized stresses for which no slip occurs; a non-associative rule with linear kinematic hardening is used to model the full-slip phase. The results of several numerical simulations for a riser of small-length, obtained with a very detailed (small-scale) non-linear finite-element model, are used to identify the parameters of the constitutive law, bridging in this way the small scale of the detailed finite-element simulations with the large scale of the beam model. The effectiveness of the proposed method is validated by the satisfactory agreement between the results of various detailed finite-element simulations for a short riser, subject to internal and external uniform pressure and uniform cyclic bending loading, with those given by the proposed constitutive law. [Copyright &y& Elsevier]

Abstract: In this paper, the first and second order derivatives of natural frequencies and buckling loads with respect to an arbitrary geometrical or physical property of a plate structure are calculated. Based on these eigenderivatives, and by using first and second order Taylor expansions, an approximate method is presented for simultaneous modification of natural frequencies and buckling loads. In the proposed method, the structural modification is considered as an inverse eigenvalue problem and calculated eigenderivatives are used for transforming this inverse problem to solving a system of algebraic equations. By solving this set of equations, the necessary changes in design parameters are calculated, in order to achieve predefined simultaneous shifts in natural frequencies and buckling loads. A plate with two boundary conditions is considered as a case study. The plate is divided into eight regions and the plate thickness in each region is considered as a design parameter. By considering several case studies, the required modification in design variables is calculated for achieving a predefined change in natural frequencies, buckling loads or simultaneous change of both. In each case, the calculated changes in design variables are implemented on the initial structure, and the changes in eigenvalues are computed using FE method. It is shown that the results from presented method compare very well with those obtained from a direct optimization procedure. [Copyright &y& Elsevier]

• Fretting fatigue behavior of 316L SS was examined under combined loading. • Combined load condition was induced using a designed fixture. • Increasing the bending contribution improves fretting fatigue life. • Increasing the grain size decreases the fretting fatigue life. • Increasing the bending contribution reduces singularity at the contact edges. • Fretting fatigue life is more sensitive to bending contribution than grain size. Fretting fatigue occurs in engineering applications where two interacting surfaces are subjected to fluctuating forces. The fluctuating forces occasionally act as a combined loading condition. The present paper investigates fretting fatigue behaviour of 316L stainless steel under such combined (i.e. tensile and bending) loading conditions. A new fixture was developed in order to apply bending and tensile loads simultaneously. Using this fixture, the effect of different bending-to-tension ratios was investigated. The results showed that increasing the contribution of bending load improves fretting fatigue life. On the other hand, increasing grain size decreases fretting fatigue life. Furthermore, sensitivity analysis based on Taguchi method revealed that fretting fatigue life is more sensitive to the increment of bending force contribution. The cyclic behavior of the specimens was assessed and finite element simulation was utilized to study the main contact parameters, including contact pressure, shear stress, and relative slip. Whilst confirming the experimental results, the simulation further showed that increasing the contribution of bending force reduces the concentration behavior at the contact edges. [ABSTRACT FROM AUTHOR]

This article presents a study on the capability of the time domain OMA method, NExT-ERA, to identify closely spaced structural dynamic modes. A survey in the literature reveals that few experimental studies have been conducted on the effectiveness of the NExT-ERA methodology in case of closely spaced modes specifically. In this paper we present the formulation for NExT-ERA. This formulation is then implemented in an algorithm and a code, developed in house to identify the modal parameters of different systems using their generated time history data. Some numerical models are firstly investigated to validate the code. Two different case studies involving a plate with closely spaced modes and a pulley ring with greater extent of closeness in repeated modes are presented. Both structures are excited by random impulses under the laboratory condition. The resulting time response acceleration data are then used as input in the developed code to extract modal parameters of the structures. The accuracy of the results is checked against those obtained from experimental tests. [ABSTRACT FROM AUTHOR]

Abstract: In this paper a modal coupling procedure is developed to improve residual modal effects based on experimentally generated data. While showing that Craig–Chang''s proposed approximation function for higher frequency modes does not compare well with the exact function, a new procedure is developed in order to re-express the dynamic residuals. For this purpose, all frequency dependent terms of the Maclaurin series in Craig–Chang''s proposed method are replaced by a general compatible function, which includes the neglected higher order terms. However, this function no longer appears in final equation of motion. Results from case studies for the discrete, continuous and experimental models are presented to verify the procedure and to show the capabilities of the proposed method. [Copyright &y& Elsevier]

In this paper, an approximate method is proposed for achieving predefined increases in the buckling threshold of a metal shear wall in order to increase its energy absorption capacity. The first and second-order derivatives of shear buckling loads of a shear wall with respect to the thickness in its different regions are calculated. Based on these eigenderivatives, and by using the first and second order Taylor expansions, the necessary change in the thickness of plate in various regions is calculated for increasing the shear buckling loads by a specific value. The presented modification algorithm is implemented for shear walls with different aspect ratios, material properties and boundary conditions. An initial sensitivity analysis is carried out for finding the regions within the shear wall where modifying the thickness has the most influence on the buckling loads. Based on the sensitivity analysis results, appropriate regions of plate are selected and the necessary modification in thickness of these regions is calculated for achieving a relatively large predefined change in shear buckling load. By simulating the post-buckling response of both initial and modified plates in a case study, the improvement in the energy absorption capability of the modified plate is also studied. [ABSTRACT FROM AUTHOR]

Abstract: Twin screw melt conditioners are used for mixing purposes and are mainly used for polymer processing. These conditioners (extruders) can be used for liquid metal processing in which liquid metal/slurry is subjected to high shear stress. This process results in grain refinement of structure. In this article, in a co-rotating twin screw melt conditioner, the solidification process of a liquid along with temperature variations of the melt with regard to the complexity of the flow has been examined. With the aid of dynamic mesh scheme, a Computational Fluid Dynamic (CFD) simulation was performed. The achieved results were in good agreement with values based on the experimental measurements. It was concluded that shearing and pouring temperatures play important roles in solidification progression and the main reason of surviving nuclei is heat dissipation from the barrel. Also, the main factor affecting the grain size is the temperature differences between the pouring and the setting temperature. It was observed that, the twin screw melt conditioner can decrease the temperature gradient and with the help of turbulence, providing appropriate conditions for formation of fine and equiaxed grains. [Copyright &y& Elsevier]

Abstract: The application of Melt Conditioner (MC) for production of high-quality DC billets and slabs at semisolid state is examined. In this technique, liquid metal is continuously fed into a twin-screw melt conditioning device, where the liquid metal or semisolid slurry is subjected to high shear rate, and then delivered continuously into a DC caster. In this paper, the feasibility of MCDC of a wrought aluminium alloy in the semi solid state has been investigated and the possibility of DC casting by MC technique below the liquidus temperature as a novel idea was explored. This idea may challenge the conventional DC casting which is normally performed above liquidus temperature. [Copyright &y& Elsevier]

Abstract: The influence of intensive shearing above liquidus on the solidification process, microstructural evolution and its effect on the final microstructure for the Al–10wt% Mg alloy were investigated and discussed thoroughly. The effect of shearing was explored and was compared with non-sheared samples. The effects of relaxation (holding melt after shearing) and relaxation with addition of grain refiners on the microstructural development were also examined. Experimental results suggest that shearing above melting point can refine the alloy structure and the level of refinement is increased by increase of shearing temperature. As regards to the relaxation time, it is inferred that shearing is a stable phenomenon and the effect of shearing is retained for a considerable time. Introduction of TiB2 results in additional, rather stable grain refining; however, with addition of TiC the obtained grain refinement effect disappears. It is suggested that dissolution and chemical reactions of TiC with the melt actually diminish the refinement effect during relaxation time. [Copyright &y& Elsevier]