IMPLEMENTATION OF STRENGTH CALCULATION SOFTWARE BASED ON THE SEMI-ANALYTICAL METHOD OF FINITE ELEMENTS

Abstract

One of the main and responsible stages of creating an apparatus for numerical analysis of structures using the finite element method is its implementation in the form of a set of programs. The principles of complex construction should take into account modern requirements for strength calculation software in modern calculation complexes. Among them, first of all, automation of the main stages of the computing process, rational use of the resources of the operating and external memory of storage devices, non-closedness in relation to the classes of problems to be solved, algorithms for the task of input data, the method of discretization and solution of systems of equations and etc. In addition, the structure of the programs should take into account the specifics of the semi-analytical method of finite elements, for which such a large experience of creating developed systems of mathematical support for the search of spatial structures has not yet been accumulated, as when using the traditional version of FEM. Considerable experience in solving finite element mechanics problems, accumulated over the past decades, has been reflected in a number of industrial commercial software complexes of domestic (LIRA, SCAD) and foreign (ANSYS, Nastran, ABAQUS) production. The developed finite element base of these software complexes allows you to obtain solutions to a wide range of problems of the mechanics of a deformable rigid body for objects of different dimensions, including for massive spatial bodies, and convenient means of input-output of information and processing of the obtained results make them very accessible to a wide circles of users and allow to display the obtained results with the maximum degree of visibility. There are also object-oriented complexes created at manufacturing enterprises and research institutes, for example at the Institute of Strength Problems of the National Academy of Sciences of Ukraine, at the Dnipropetrovsk National University.
 The conducted research will determine the most optimal ways of solving the listed problems from the point of view of computing costs and complexity of implementation, as well as outline the circle of unsolved issues.