• Steel and Composite Structures
• /
• v.51 no.2
• /
• pp.203-223
• /
• 2024

This study shows functionally graded material structural topology optimization under buckling constraints. The SIMP (Solid Isotropic Material with Penalization) material model is used and a method of moving asymptotes is also employed to update topology design variables. In this study, the quadrilateral element is applied to compute buckling load factors. Instead of artificial density properties, functionally graded materials are newly assigned to distribute optimal topology materials depending on the buckling load factors in a given design domain. Buckling load factor formulations are derived and confirmed by the resistance of functionally graded material properties. However, buckling constraints for functionally graded material topology optimization have not been dealt with in single material. Therefore, this study aims to find the minimum compliance topology optimization and the buckling load factor in designing the structures under buckling constraints and generate the functionally graded material distribution with asymmetric stiffness properties that minimize the compliance. Numerical examples verify the superiority and reliability of the present method.

• Advances in aircraft and spacecraft science
• /
• v.10 no.6
• /
• pp.521-543
• /
• 2023

Reynolds-averaged Navier-Stokes (RANS) models are extensively employed in industrial settings for the purpose of simulating intricate fluid flows. However, these models are subject to certain limitations. Notably, disparities persist in the Reynolds stresses when comparing the RANS model with high-fidelity data obtained from Direct Numerical Simulation (DNS) or experimental measurements. In this work we propose an approach to mitigate these discrepancies while retaining the favorable attributes of the Menter Shear Stress Transport (SST) model, such as its significantly lower computational expense compared to DNS simulations. This strategy entails incorporating an explicit algebraic model and employing a neural network to correct the turbulent characteristic time. The imposition of realizability constraints is investigated through the introduction of penalization terms. The assimilated Reynolds stress model demonstrates good predictive performance in both in-sample and out-of-sample flow configurations. This suggests that the model can effectively capture the turbulent characteristics of the flow and produce physically realistic predictions.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.3
• /
• pp.159-165
• /
• 2021

This study introduces a topology optimization method for the simultaneous design of macro-scale structural configuration and unit structure variation to ensure effective heat conduction. Shape changes in the unit structure depending on its location within the macro-scale structure result in micro- as well as macro-scale design and enable better performance than using isotropic unit structures. They result in functionally graded composite structures combining both configurations. The representative volume element (RVE) method is applied to obtain various thermal conductivity properties of the multi-material based unit structure according to its shape change. Based on the RVE analysis results, the material properties of the unit structure having a certain shape can be derived using machine learning. Macro-scale topology optimization is performed using the traditional solid isotropic material with penalization method, while the unit structures composing the macro-structure can have various shapes to improve the heat conduction performance according to the simultaneous optimization process. Numerical examples of the thermal compliance minimization issue are provided to verify the effectiveness of the proposed method.

• Smart Structures and Systems
• /
• v.11 no.3
• /
• pp.241-259
• /
• 2013

This paper presents the optimum load-speed diagram evaluation for a linear micromotor, including multitude cantilever piezoelectric bimorphs, briefly. Each microbeam in the mechanism can be actuated in both axial and flexural modes simultaneously. For this design, we consider quasi-static and linear conditions, and a relatively new numerical method called the smoothed finite element method (S-FEM) is introduced here. For this purpose, after finding an optimum volume fraction for piezoelectric layers through a standard numerical method such as quadratic finite element method, the relevant load-speed curves of the optimized micromotor are examined and compared by deterministic topology optimization (DTO) design. In this regard, to avoid the overly stiff behavior in FEM modeling, a numerical method known as the cell-based smoothed finite element method (CS-FEM, as a branch of S-FEM) is applied for our DTO problem. The topology optimization procedure to find the optimal design is implemented using a solid isotropic material with a penalization (SIMP) approximation and a method of moving asymptotes (MMA) optimizer. Because of the higher efficiency and accuracy of S-FEMs with respect to standard FEMs, the main micromotor characteristics of our final DTO design using a softer CS-FEM are substantially improved.

• Journal of Ocean Engineering and Technology
• /
• v.24 no.1
• /
• pp.153-160
• /
• 2010

This paper deals with the topology optimized design of the riser support structures for floating production storage and offloading units (FPSOs) under global and local loading conditions. For a preliminary study and validation of the numerical approach, a simplified plate under static loading is first evaluated with the representative topology optimization methods, the Homogenization Design Method (HDM) and Density Method (DM) or Simple Isotropic Material with Penalization (SIMP). In the context of the corresponding riser support structures, the design problem is formulated such that structure shapes based on design domain variables are determined by minimizing the compliance subject to a mass target, considering the stress criterion. An initial design model is generated based on an actual FPSO riser support configuration. The topology optimization results present improved design performances under various loading conditions, while staying within the allowable limit of the offshore area.

• The Korean Journal of Applied Statistics
• /
• v.29 no.6
• /
• pp.1147-1154
• /
• 2016

In this paper, we numerically compare two penalized least square methods, the ${\ell}_0$-penalized method and the fused lasso regression (FLR, ${\ell}_1$ penalization), in finding multiple change points of a signal. We find that the ${\ell}_0$-penalized method performs better than the FLR, which produces many false detections in some cases as the theory tells. In addition, the computation of ${\ell}_0$-penalized method relies on dynamic programming and is as efficient as the FLR.

• Journal of computational fluids engineering
• /
• v.22 no.1
• /
• pp.51-58
• /
• 2017

A penalized VIC method offers an efficient hybrid particle-mesh algorithm to simulate an incompressible viscous flow passing a solid body in an infinite domain. In this manner, the computational domain needs to be restricted to a relatively small region to reduce computational cost which would be very high in case of using a large domain. In this paper, we present how to dispose of far-field particles to avoid an unnecessarily large computational domain. The present approach constraints expansion of the domain and thus prevents the incremental computational cost. To validate the numerical approach, a flow around an impulsively started sphere was simulated for Reynolds numbers of 100 and 1000.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.23 no.1
• /
• pp.84-91
• /
• 2013

Ultrasonic flip-chip bonding needs a precise bonding tool which delivers ultrasonic energy into chip bumps effectively to use the selected resonance mode and frequency of the horn structure. The bonding tool is excited at the resonance frequency and the input and output ports should locate at the anti-nodal points of the resonance mode. In this study, we propose new design method with topology optimization for ultrasonic bonding tools. The SIMP(solid isotropic material with penalization) method is used to formulate topology optimization and OC(optimal criteria) algorithm is adopted for the update scheme. MAC(modal assurance criterion) tracking is used for the target frequency and mode. We fabricate two prototypes of ultrasonic tools which are based on 3D optimization models after reviewing 2D and 3D topology optimization results. The prototypes are satisfied with the ultrasonic frequency and vibration amplitude as the ultrasonic bonding tools.

• International Journal of Aeronautical and Space Sciences
• /
• v.14 no.4
• /
• pp.324-333
• /
• 2013

The purpose of this paper is to improve the efficiency of multi-objective topology optimization using a genetic algorithm (GA) with bar-system representation. We proposed a new GA using an elite initial population obtained from a Solid Isotropic Material with Penalization (SIMP) using a weighted sum method. SIMP with a weighted sum method is one of the most established methods using sensitivity analysis. Although the implementation of the SIMP method is straightforward and computationally effective, it may be difficult to find a complete Pareto-optimal set in a multi-objective optimization problem. In this study, to build a more convergent and diverse global Pareto-optimal set and reduce the GA computational cost, some individuals, with similar topology to the local optimum solution obtained from the SIMP using the weighted sum method, were introduced for the initial population of the GA. The proposed method was applied to a structural topology optimization example and the results of the proposed method were compared with those of the traditional method using standard random initialization for the initial population of the GA.

• ETRI Journal
• /
• v.27 no.6
• /
• pp.747-758
• /
• 2005

The problem of selection among competing models has been a fundamental issue in statistical data analysis. Good fits to data can be misleading since they can result from properties of the model that have nothing to do with it being a close approximation to the source distribution of interest (for example, overfitting). In this study we focus on the preference among models from a family of polynomial regressors. Three decades of research has spawned a number of plausible techniques for the selection of models, namely, Akaike's Finite Prediction Error (FPE) and Information Criterion (AIC), Schwartz's criterion (SCH), Generalized Cross Validation (GCV), Wallace's Minimum Message Length (MML), Minimum Description Length (MDL), and Vapnik's Structural Risk Minimization (SRM). The fundamental similarity between all these principles is their attempt to define an appropriate balance between the complexity of models and their ability to explain the data. This paper presents an empirical study of the above principles in the context of model selection, where the models under consideration are univariate polynomials. The paper includes a detailed empirical evaluation of the model selection methods on six target functions, with varying sample sizes and added Gaussian noise. The results from the study appear to provide strong evidence in support of the MML- and SRM- based methods over the other standard approaches (FPE, AIC, SCH and GCV).