Preface
In 1979 Professors R.W. Lewis and K. Morgan of the University of Wales,Swansea organized the first International Conference on Numerical Methods in Thermal Problems. A conference viewed as a foundation point in the field of numerical heat transfer. In 2004 Professor Lewis organized a symposium on numerical heat transfer at the European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS) held in Jyväskylä Finland,July 2004. The papers presented at that meeting, fully demonstrated the extent of the advance of numerical heat transfer in the 25 years since the initial meeting in Swansea. This special issue of the International Journal of Numerical Methods for Heat & Fluid Flow illustrates the nature of these advances by presenting a selection of extended versions of the papers presented at ECCOMAS.
A major accomplishment of numerical heat transfer has been to provide tools for the simulation and analysis of relevant industrial process. The first two papers in this special issue represent advances in this arena. The first,appropriately from Professor Lewis’s research group in Swansea – A finite element model of the squeeze casting process, R.W. Lewis, Eligiusz W. Postek, Zhiqiang Han, David T. Gethin, and Rajesh S. Ransing – presents a truly comprehensive model of a metal casting process. This paper provides a road map on how mechanical, fluid flow, and heat transfer phenomena can be coupled into an effective process simulation. A major component in this multi-physics model is the tracking of the metal free surface during the mold filling, an issue that is common to nearly all casting models and the main focus of the second paper – Finite element modeling of the lost foam casting process tackling back-pressure effects, Guillaume Houzeaux and Ramon Codina. In order to tack the metal/form interface these authors develop and demonstrate a novel level set approach operating in conjunction with a fixed mesh ALE. A feature in this work is the accounting of the back-pressure arising from the combustion of the foam.
Both the models introduced in the first two papers are large-scale problems. The ability to numerically resolve problems of this nature rests to a large extent on advances in the numerical treatment of boundary conditions and the development of efficient solvers; the topics of the next two papers in this special issue. In the third paper – Linear tetrahedral finite elements for thermal shock problems, Víctor D. Fachinotti and Michel Bellet –develop a finite element treatment that allows for a direct treatment of a boundary condition consisting of a very rapid transient change in temperature (a thermal shock). In a basic approach, without modification, such a condition could result in the prediction of oscillations in the temperature field. This paper proposes a method that uses a modified conductivity matrix over an initial time interval. An approach that allows for solutions based on linear elements and explicit time integration. Although not specifically focused on a heat transfer topic the fourth paper in the issue – On iterative solvers for non-Newtonian flow equations, Oleg Iliev, Joachim Linn, Mathias Moog, Dariusz Niedziela, and Vadimas Starikovicius – considers the important topic of evaluating efficient iterative solvers. As numerical heat transfer problems become larger the need for efficient solvers becomes more critical and this paper evaluates a number of high end iterative solvers when applied to problems that contain physical features usually not accounted for in basic test problems.
In addition to the application to large-scale heat transfer problems and the refinement and improvement of existing schemes the field of numerical heat transfer also includes research directed at developing novel numerical methods and the application of numerical heat transfer methods in solving problems in other areas of science and engineering. The last two papers in this special issues look at contributions in these two areas. The fifth paper –Meshless local radial basis function collocation method for convective-diffusive solid-liquid phase change problems, Robert Vertnik and Boidar Šarler– presents the development of a state of the art meshless method. In the general field of computational mechanics this relatively new discretization method is ideal for handling problems with transient discontinuities. In the paper presents here the authors develop a meshless method for the well known moving boundary problem of solid-liquid phase change. The sixth and last paper– An enthalpy method for moving boundary problems on the earths surface preface, V.R. Voller, J. B. Swenson, W. Kim and C. Paola – reports from work by the National Center for Earth Surface Dynamics (NCED). At first glance the topic – the formation of sedimentary basins – is far removed from heat transfer topics. The numerical tools used in this work, however, are essentially slight modifications of the standard numerical heat transfer tools used to model model the classic Stefan melting problem.
The papers in this special issue highlight many of the important active areas in numerical heat transfer research – simulation of multi-physics problems, the refinement of treatments for boundary conditions, the identification of efficient solvers, the development of new approaches, and the extension of numerical heat transfer techniques into other fields. Together they provide a window on the advances made in numerical methods for thermal problems since the groundbreaking conference organized by Professors Lewis and Morgan in 1979.
Vaughan Voller
