Search published articles

Showing 3 results for Roodsarabi

M. Roodsarabi, M. Khatibinia , S. R. Sarafrazi,
Volume 6, Issue 3 (9-2016)

This study focuses on the topology optimization of structures using a hybrid of level set method (LSM) incorporating sensitivity analysis and isogeometric analysis (IGA). First, the topology optimization problem is formulated using the LSM based on the shape gradient. The shape gradient easily handles boundary propagation with topological changes. In the LSM, the topological gradient method as sensitivity analysis is also utilized to precisely design new holes in the interior domain. The hybrid of these gradients can yield an efficient algorithm which has more flexibility in changing topology of structure and escape from local optimal in the optimization process. Finally, instead of the conventional finite element method (FEM) a Non–Uniform Rational B–Splines (NURBS)–based IGA is applied to describe the field variables as the geometry of the domain. In IGA approach, control points play the same role with nodes in FEM, and B–Spline functions are utilized as shape functions of FEM for analysis of structure. To demonstrate the performance of the proposed method, three benchmark examples widely used in topology optimization are presented. Numerical results show that the proposed method outperform other LSMs.

M. Khatibinia, M. Roodsarabi, M. Barati,
Volume 8, Issue 2 (8-2018)

This paper presents the topology optimization of plane structures using a binary level set (BLS) approach and isogeometric analysis (IGA). In the standard level set method, the domain boundary is descripted as an isocountour of a scalar function of a higher dimensionality. The evolution of this boundary is governed by Hamilton–Jacobi equation. In the BLS method, the interfaces of subdomains are implicitly represented by the discontinuities of BLS functions taking two values 1 or −1. The subdomains interfaces are represented by discontinuities of these functions. Using a two–phase approximation and the BLS approach the original structural optimization problem is reformulated as an equivalent constrained optimization problem in terms of this level set function. For solving drawbacks of the conventional finite element method (FEM), IGA based on a Non–Uniform Rational B–Splines (NURBS) is adopted to describe the field variables as the geometry of the domain. For this purpose, the B–Spline functions are utilized as the shape functions of FEM for analysis of structure and the control points are considered the same role with nodes in FEM. Three benchmark examples are presented to investigate the performance the topology optimization based on the proposed method. Numerical results demonstrate that the BLS method with IGA can be utilized in this field.
M. Khatibinia, M. Roodsarabi,
Volume 10, Issue 3 (6-2020)

The present study proposes a hybrid of the piecewise constant level set (PCLS) method and isogeometric analysis (IGA) approach for structural topology optimization. In the proposed hybrid method, the discontinuities of PCLS functions is used in order to present the geometrical boundary of structure. Additive Operator Splitting (AOS) scheme is also considered for solving the Lagrange equations in the optimization problem subjected to some constraints. For reducing the computational cost of the PCLS method, the Merriman–Bence–Osher (MBO) type of projection scheme is applied. In the optimization process, the geometry of structures is described using the Non–Uniform Rational B–Splines (NURBS)–based IGA instead of the conventional finite element method (FEM). The numerical examples illustrate the efficiency of the PCLS method with IGA in the efficiency, convergence and accuracy compared with the other level set methods (LSMs) in the framework of 2–D structural topology optimization. The results of the topology optimization reveal that the proposed method can obtain the same topology in lower number of convergence iteration.

Page 1 from 1     

© 2024 CC BY-NC 4.0 | Iran University of Science & Technology

Designed & Developed by : Yektaweb