Seog-Young Han

Application of artificial bee colony algorithm to topology optimization for dynamic stiffness problems

The artificial bee colony algorithm (ABCA) was first adopted in topology optimization for dynamic problems. The objective was to obtain a structure with the highest fundamental natural frequency in a certain amount of material, based on the contributed structural sensitivity of each element calculated by the waggle index and eigenvalue. The waggle index update rule, evaluation method of fitness values, and changing filtering size scheme are suggested for obtaining a stable and robust optimal topology based on the ABCA. Examples are provided to examine the applicability and effectiveness of the ABCA compared to bi-directional evolutionary structural optimization (BESO). The following conclusions are obtained through the results of examples based on the ABCA; (1) the ABCA, using the three suggested methods, is very applicable and effective in topology optimization for obtaining a stable and robust optimal layout. (2) It is found that the natural frequencies of the ABCA are always higher than those of the BESO, and average convergence rates of the ABCA are similar or faster than those of the BESO. (3) The optimal topology from the ABCA is nearly obtained in a half stage of the convergence iteration, since volume constraint is applied from the beginning.

A modeling approach to energy savings of flying Canada geese using computational fluid dynamics

A flapping flight mechanism of the Canada goose (Branta canadensis) was estimated using a two-jointed arm model in unsteady aerodynamic performance to examine how much energy can be saved in migration. Computational fluid dynamics (CFD) was used to evaluate airflow fields around the wing and in the wake. From the distributions of velocity and pressure on the wing, it was found that about 15% of goose flight energy could be saved by drag reduction from changing the morphology of the wing. From the airflow field in the wake, it was found that a pair of three-dimensional spiral flapping advantage vortices (FAV) was alternately generated. We quantitatively deduced that the optimal depth (the distance along the flight path between birds) was around 4m from the wing tip of a goose ahead, and optimal wing tip spacing (WTS, the distance between wing tips of adjacent birds perpendicular to the flight path) ranged between 0 and -0.40m in the spanwise section. It was found that a goose behind can save about 16% of its energy by induced power from FAV in V-formation. The phase difference of flapping between the goose ahead and behind was estimated at around 90.7° to take full aerodynamic benefit caused by FAV.

APPLICATION OF THE GROWTH-STRAIN METHOD FOR SHAPE OPTIMIZATION OF FLOW SYSTEMS

In this study, the growth-strain method was used for shape optimization of flow systems. It optimizes a shape by making a distributed parameter such as dissipation energy uniform in a flow system. In order to overcome the instability that occurred in the numerical analysis by the growth-strain method, the equation of bulk strain has been modified. And the distributed parameters were variously established in this study. By comparing the optimized shapes with the known optimal shapes for two flow systems, it is confirmed that the modified growth-strain method is very efficient and practical in shape optimization of flow systems.

The speed of sound through trabecular bone predicted by Biot theory

Cancellous bone is a highly porous material filled with fluid. The mechanical properties of cancellous bone determine whether the bone is normal or osteoporotic. Wave propagation can be used to measure the elastic constants of cancellous bone. Recently, poroelasticity theory has been used to predict the elastic constants of cancellous bone from the wave velocities. In this study, it is shown that the fast wave, predicted by the Biot theory, corresponds to the wave penetrating the trabeculae, while the slow wave is determined by the interaction between the trabeculae and the fluid. The trabecular shape does not affect the wave velocity significantly when using the variable, which is determined by the microstructure, and the slow wave velocity decreases after the porosity reaches 80%.

Shape Control of Alloy Steel Rolled by Sendzimir Mill

In order to improve quarter waves occurred in the wide and thin gauged alloy steel rolled by 20-high sendzimir mill, a computer simulation based on the divided element method and an actual cold rolling experiment were carried out. Quarter waves were simulated by elastic deformation analysis of rolls considering bending deformation of back up rolls and the effect of control actuators on controllability of quarter waves were analyzed. Computer simulation showed that control actuators such as shifting of the 1st intermediate roll and crown adjustment of As-U-Roll in back up rolls were not effective to control quarter waves and that changing taper mode (both length and magnitude) at the barrel-end taper radius of the 1st intermediate roll was rather very effective. From an actual rolling experiment it was verified that quarter waves could be reduced remarkerably by changing taper mode of the 1st intermediate roll.

Application of artificial bee colony algorithm for structural topology optimization

Artificial bee colony algorithm (ABCA) as one of swarm intelligence methods and finite element analysis are first adopted for structural topology optimization. The objective of the paper is to examine the effectiveness and applicability of the suggested ABCA in structural topology optimization. This paper describes considerable modifications made to the ABCA in order to solve discrete topology optimization problems. The desirable conclusions are obtained through the results of the examples based on the suggested ABCA.

Parameter Optimization of a Micro-Static Mixer Using Successive Response Surface Method

In this study, parameter optimization of micro-static mixer with a cantilever beam was accomplished for maximizing the mixing efficiency by using successive response surface approximations. Variables were chosen as the length of cantilever beam and the angle between horizontal and the cantilever beam. Sequential approximate optimization method was used to deal with both highly nonlinear and non-smooth characteristics of flow field in a micro-static mixer. Shape optimization problem of a micro-static mixer can be divided into a series of simple subproblems. Approximation to solve the subproblems was performed by response surface approximation, which does not require the sensitivity analysis. To verify the reliability of approximated objective function and the accuracy of it, ANOVA analysis and variables selection method were implemented, respectively. It was verified that successive response surface approximation worked very well and the mixing efficiency was improved very much comparing with the initial shape of a micro-static mixer.

Dynamic topology optimization based on ant colony optimization

A modified ant colony optimization algorithm implementing a new definition of pheromone and a new cooperation mechanism between ants is presented in this paper. The study aims at improving the suitability and computational efficiency of the ant colony optimization algorithm in dynamic topology optimization problems. The natural frequencies of the structure must be maximized yet satisfying a constraint on the final volume. Optimization results obtained in three test cases indicate that modified ACO is more efficient and robust than ACO in solving dynamic topology optimization problems.

Optimization of the Micro-Mixer with an Oscillating Stirrer using the Kriging Method

Microfluidic analysis of an active micro-mixer with an oscillating stirrer in a straight micro-channel was performed. An optimum design for an active micro-mixer with an oscillating stirrer was performed based on the kringing method. It was found that the mixing index of the optimal design was improved by up to 70.12% compared to that of the initial design.

Axiomatic Design of a Micromanipulator using Displacement Amplifier

Micromanipulator is a device that manipulates an object with high precision. Generally, a parallel-type robot has inherently higher precision than a serial-type robot. In most cases, the use of flexure hinge mechanisms is the most appropriate approach to micromanipulators. The micromanipulator is basically required that have high natural frequency and sufficient workspace. However, previous designs are hard to satisfy the required workspace and natural frequency, simultaneously, because the previous micromanipulators are coupled designs. Therefore, this paper suggests a new design parameter as displacement amplifier and new design procedure based on semi-coupled design in axiomatic design. As a consequence the spatial 3-DOF micromanipulator which is chosen as an exemplary device has natural frequency of 500Hz and workspace of . To investigate the effectiveness of the displacement amplifier, simulation and experiment are performed.