Boundary element analysis of nonhomogeneous biharmonic phenomena (Q4023635)

The objective of this text is to present a more-or-less self-contained treatment of boundary integral analysis of the nonhomogeneous biharmonic equation, and to present certain refinements that make the method more accurate as well as computationally more efficient. The necessary boundary integral relationships for very general cases are derived, and detailed analyses and implementation strategies for various element types are described. With the Overhauser element and the domain fanning system, a wide variety of problems have been solved with a precision and ease that are unprecedented in previous works. These problems include both nonhomogeneous and nonlinear variations of the basic biharmonic problem. The nonlinear problems are handled by incorporating an iterative solution technique which takes advantage of an interpolating domain map storage scheme. Precise solutions have been obtained for a diverse group of problems in which the nonlinearities involve both the field variable and its various derivatives. The example problems, presented in Chapter 5, not only corroborate this claim of accuracy and versatility, but also imply the formulation's ability to solve similar problems of this nature. Finally, a detailed computer program suitable for solving a general two- dimensional nonhomogeneous biharmonic problem has been presented. The program includes constant, linear, quadratic, and Overhauser elements, as well as the fanning domain integrator for handling general domain terms. This program may be used to solve any of the single iteration problems described in Chapter 5. Several conclusions are evident from the research portion of this work, and it is easy to extrapolate them into general statements that carry over into other areas of BEM. Significant progress is made in analyzing nonlinear forms of equations involving biharmonic operators. The iterative solution technique employing the interpolation map storage scheme proves to be quite efficient in solving the form of the equation in which the nonhomogeneous term is a function of the field variables and their various partial derivatives. A variety of examples presented in Chapter 5 demonstrate the ability of this method for solving extremely complex plate bending and creeping flow problems for which solutions by analytical means are virtually intractable. The computer program presented is probably the only published one for the analysis of the general biharmonic equation. The art and science of boundary elements are no longer in their infancy, but they are still in their childhood. Brebbia mentioned in 1985 that the boundary element research community should learn from the mistakes of the early finite element researchers, and attempt not to go too far afield without quantifying the problems and errors associated with the present state of the art. In order for the method to grow properly, it is necessary that all BEM techniques here or elsewhere should be subjected to the utmost scrutiny. That is why we look forward to the present work being made obsolete, and replaced in the foreseeable future, if not by ourselves, by someone who has been sufficiently inspired and interested to take the work to further heights.