Adaptive structural design - Integration of computational tools for CAD, adaptive FEM and optimization
Licentiatavhandling, 1999

Background Construction is one of the largest industries with major impact on economy, environment and society. Improved structural design is important since the decisions made during the design process strongly influence all later stages, i.e. construction, operation, maintenance, repair and disassembly, and hence have major impact on the total economy. Optimization methods, FEM (Finite Element Method) for numerical solution of mathematical models and parametric 3D CAD-technique (Computer Aided Design) for geometry description in an overall high performance computing process are increasingly used in mechanical engineering. In civil engineering the application of these methods is not fully exploited. Aim The aim of this thesis is to contribute in the development of tools for rational structural design based on mathematics, optimization, adaptive computation and integration of methods into an “adaptive structural design” loop. Method The research method has a development and synthesis character. Ideas and methods are continuously implemented, incorporated and tested on realistic cases; e.g. column supported slabs and a common type of roadway bridge. Results The thesis has presented the notion of “adaptive structural design”: 1. Structural design has been idealized as an optimization problem. 2. Programs for optimization, geometry description (CAD) and structural analysis (FEM) have been integrated to solve the optimization problem. 3. Discretization adaptivity has been tested and mathematical model adaptivity has been proposed as means to adapt the computational effort to the accuracy needed in the current phase of the design process. The tests have indicated that even for repeatedly built, standard type of structures economical savings can be utilized by improved computational methods in structural design.


structural mechanics







Per Kettil

Institutionen för byggnadsmekanik



Publication - Chalmers University of Technology, Department of Structural Engineering: 99:1