Shape Optimization and Parametric Modeling

Publications

Springback Compensation in Sheet Metal Forming Using a Successive Response Surface Method

This paper explores an iterative algorithm for springback compensation in sheet metal stamping. It applies an optimization method based on a successive response surface scheme to iteratively modify tool geometry using a parametric model, demonstrating a more effective and less costly alternative to traditional trial-and-error methods.

Refined 3D Finite Element Modeling of Partially Restrained Connections Including Slip

A.M. Citipitioglu, R.M. Haj-Ali, D.W. White

This paper introduces a sophisticated parametric 3D FEM approach for analyzing partially-restrained bolted steel beam-column connections. It examines how varying geometrical and material parameters, including critical aspects like bolt pretension and slip, influence the connection's complex nonlinear behavior. The developed versatile modeling technique, which employs parametric meshing, enables accurate prediction and extensive parametric studies for a broad spectrum of connection designs.

Stochastic Analysis of Highly Non-Linear Structures

W. Roux, N. Stander, F. Günther, H. Müllerschön

The paper explores how shape optimization and parametric modeling techniques can be used to analyze and mitigate variability in highly non-linear structural systems. By applying response surface metamodels and stochastic analysis, it differentiates between design-driven and instability-induced changes. These insights are crucial for reliably modeling, optimizing, and visualizing complex structural responses.

Optimization of Draw-In for an Automotive Sheet Metal Part

This paper details an optimization study for draw-in in automotive sheet metal parts, using a blend of response surface methodology (RSM) and space mapping (SM) to calibrate a finite element model with physical tests. The methodology effectively translates the equivalent draw bead restraining force from the simulation into an accurate physical draw bead geometry, demonstrating an efficient way to automate a process traditionally done manually.

New Developments in LS-OPT - Robustness Studies

Ken Craig, Willem Roux, Nielen Stander

Presents a robustness study for the optimization of shell buckling by incorporating random geometric imperfections modeled using Karhunen-Loève expansions. The process uses LS-OPT and Monte Carlo analysis to perform stochastic optimization and create a more reliable design that can handle these imperfections.