The finite element method in structural mechanics. Principles and practice of design of field-consistent element for structural and solid mechanics (Q1904031)

The main objective of the book is the design of robust and accurate finite elements for applications in structural mechanics. The base of the finite element method is omitted in the contents since it can be found widely treated in numerous textbooks. The main emphasis of the book is to establish the process by which continuum structural behaviour is replaced by a discretized description so that these can be coded as mechanics algorithms. The early efforts to do this were founded on engineering principles proceeding mostly from heuristic judgement. However, considerable critical and analytical studies in recent years have shown that it is possible to establish these methods on a more scientific basis. This book therefore sets this aspect of the use of the finite element method against such principles. The author hopes that the book will serve as an up-to-date account of the state-of-the-art of the theory of finite element development and its critical assessment. In the introduction, the basic concepts are presented to understand why some discretized descriptions fail while others succeed. First, the essential feature of a simple constrained minimal problem is discussed. Then the differential equations of Timoshenko beam are derived to show how constraints are generated in such a model. A Ritz type approximation shows how to link rank and nonsingularity of stiffness matrix to the consistency of the strain field approximation. This permits to discuss the concept of consistency and the nature of errors that appear in the displacement type FEM models. The second chapter (Shear flexible beam elements) deals with the Timoshenko beam element. Shear locking is identified here, and a mathematically rigorous error model is devised. The next chapter (Simple curved beam elements) proceeds to the identification of what is called the membrane locking phenomenon. Then it is shown what happens if the mapping from the natural coordinate system to the Cartesian system is no longer linear (Non-uniform mapping and quadratic beam element), and what are the effects of a variationally incorrect formulations. Next the general curved beam element, plane stress elements, plate elements, brick elements and shell elements are studied. Finally, the author investigates the effects of constraints appearing due to the physical regimes considered, e.g. vanishing shear strains in the Timoshenko beam or Mindlin plate, vanishing membrane strains in curved beams and shells etc. He noted that the minimum total potential is inadequate to explain these difficulties, and that more general variational theorems like the Hellinger-Reissner principle were needed. The specific area to which the book is devoted makes it interesting for all working in the field of the finite element method and generally -- in the discrete modelling. However, the fundamental skills in FEM are required by the reader. Theoretical considerations are widely illustrated by numerical examples. Indications and comparison of commonly used finite element models allow to select proper models in practice.