The objective of this thesis was to assess the validity of the fluid-structure interaction (FSI) facilities in LS-DYNA for the analysis of highly deformable structures interacting with flowing viscous fluids. The collapsible tube experiment was chosen as a validation tool for FSI since its three-dimensional computational modelling would have been impossible if the viscous internal fluid flow were not considered. An explicit three-dimensional finite element model of a collapsible-tube was constructed and solved using LS-DYNA. The fully coupled model included internal fluid flow; external, inlet and outlet pressures; tube wall tension; pre-stressing; and contact. The finite element boundary conditions were taken as the recorded values of flow rate and pressure from a standard collapsible-tube experiment for both steady and unsteady flows. The predicted tube geometry in the steady LS-DYNA model showed good agreement with the experiment for operating points in the highly compliant region of the pressure-flow characteristic curve. The comparative position of the pinch at the outlet end differed by only 5.6% of the outlet diameter in the worst case. This analysis represents an advance on other published work in that previously no comparison with experiments have been drawn for FSI models involving high Reynolds number flowing viscous fluids interacting with highly deformable three dimensional structures. This analysis successfully made that comparison and the experimental and computational results have combined to form a more detailed picture of the collapsible-tube phenomenon by including detailed stress results of the tube walls and views of the internal fluid flow. The collapsible tube model exhibited uncertainty errors due to the use of a coarser than desirable mesh and a reduced fluid speed of sound. Although both these approximations caused significant error in the model both were necessary in order to achieve acceptable solution times. Because of these errors a thorough quantitative validation could not be achieved although LS-DYNA has been proven to be qualitatively accurate. Increases in computing speed are required before thorough quantitative validation of FSI can be achieved by comparison with the collapsible tube experiments.

This thesis is concerned with a theoretical and experimental study of the hydrodynamics and heat transfer characteristics of viscous fluids flowing in tubes under laminar conditions. Particular attention has been given to the effects of the rheological properties and their variation with temperature. A review of problems of this type showed that in spite of the many potential applications of the results in a wide range of industries the subject had not been well developed and further work is justified in order to fill some of the gaps in our knowledge. The early part of the thesis considers the justification of the work in this way and sets down the scope and objectives. A computer progracune was then developed to allow the governing equations of the problem to be solved numerically to give the velocity and temperature profiles and pressure drop for both heating and cooling conditions. The results were also presented in the form of Nusselt numbers as a function of the Graetz numberp since this form is useful for engineering design purposes. The validity of the predictions were then checked by a programme of experimental work. Temperature and velocity profiles have been measured in order to provide a more severe test of the theory than could be imposed by the measurement of overall heat transfer rates. A combined thermocouple probe/Pitot tube was developed to allow simultaneous measurements of velocity and temperature to be made. A Newtonian oil and two non-Newtonian Carbopol solutions were studied. This is the first time that velocity and temperature profiles have been measured for non-Newtonian fluids in this type of situation. The results of the work heve shown that (a) the velocity and temperature profiles and pressure drops are greatly affected by the temperature dependence of the rheological properties and since real viscous fluids are normally very temperature-sensitive it is important that this effect is properly taken into account. (b) the engineering design correlations commonly used for the prediction of heat transfer coefficients can be seriously in error, especially for cooling conditions and when non-Nevitonian fluids are being considered. (c) a mathematical model can be developed which accurately describes all the phenomena and gives predictions which are very close to those observed experimentally. An important objective was to develop more accurate engineering design correlations for non-isothermal pressure drop and heat transfer rates.