• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2003.11a
• /
• pp.389-394
• /
• 2003

In this paper, a shape optimization technique is applied to design of an air-conditioner mounting bracket. The mounting bracket is a structural component of an engine, on which bolts attach an air-conditioner compressor. The air-conditioner mounting bracket has a large portion of weight among the engine components. To reduce weight of the bracket, the shape is optimized using a finite element software. The compressor assembly, composed of a compressor and a bracket is modeled using finite elements. An objective function for the shape optimization of the bracket is the weight of the bracket. Two design constraints on the bracket are the first resonant frequency of the compressor assembly and the fatigue life of the bracket. The design variables are the shape of the bracket including thickness profiles of the front and back surfaces of the bracket, radius of outer bolt-holes, and side edge profiles. The coordinates of the FE nodes control the shape parameters. Optimal shapes of the bracket are obtained by using SOL200 of MSC/NASTRAN.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2003.05a
• /
• pp.501-509
• /
• 2003

A method of identifying structural parameters such as stiffness and damping coefficients at interfacial points of vibro-acoustic systems is suggested using an optimization technique. To identify the parameters using a numerical optimization algorithm, cost functions are defined. The cost function should be zero at the correct parameter values. To minimize the cost functions using an optimization technique, a design sensitivity analysis procedure is developed in the framework of the multi-domain FRF-based substructuring method. As a numerical example, a ladder-like structure problem is introduced. With known parameter values and different initial guesses of the parameters, convergence characteristics to the exact value are compared for the three cost functions. Investigating the contours of the cost functions, we find the first cost function has the largest convergent region to the correct value. As another practical problem, stiffnesses of engine mounts and bushings in a passenger car are identified. The numerical examples show that the proposed method is efficient and accurate even when applied to realistic problems.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.37 no.2
• /
• pp.131-138
• /
• 2009

Analytical modal verification is considered as the process to provide an acceptable description of the subject structure's behaviour. In general, results of original analytical model are different with actual structure results to uncertainty like non-linearity of material, boundary and modified shape, etc. In this paper, the dynamic model of glider's wing is correlated with static deformation and vibration test results by goal-attainment method, multi-objects optimization technique. The structural responses are predicted by using finite element method and optimization is carried out by using the SQP(sequential quadratic programming) method which is widely used in the constrained nonlinear optimization problem. The MAC(Modal Assurance Criterion) is used to modify the mode shapes and quantify the similarity.

• Smart Structures and Systems
• /
• v.24 no.6
• /
• pp.733-744
• /
• 2019

Wind speed forecasting is critical for a variety of engineering tasks, such as wind energy harvesting, scheduling of a wind power system, and dynamic control of structures (e.g., wind turbine, bridge, and building). Wind speed, which has characteristics of random, nonlinear and uncertainty, is difficult to forecast. Nowadays, machine learning approaches (generalized regression neural network (GRNN), back propagation neural network (BPNN), and extreme learning machine (ELM)) are widely used for wind speed forecasting. In this study, two schemes are proposed to improve the forecasting performance of machine learning approaches. One is that optimization algorithms, i.e., cross validation (CV), genetic algorithm (GA), and particle swarm optimization (PSO), are used to automatically find the optimal model parameters. The other is that the combination of different machine learning methods is proposed by finite mixture (FM) method. Specifically, CV-GRNN, GA-BPNN, PSO-ELM belong to optimization algorithm-assisted machine learning approaches, and FM is a hybrid machine learning approach consisting of GRNN, BPNN, and ELM. The effectiveness of these machine learning methods in wind speed forecasting are fully investigated by one-year field monitoring data, and their performance is comprehensively compared.

• Structural Engineering and Mechanics
• /
• v.81 no.6
• /
• pp.677-689
• /
• 2022

Generally, the goal of seismic retrofit design of an existing structure using energy dissipation devices is to determine the optimum design parameters of a retrofit device to satisfy a specified limit state with minimum cost. However, the presence of multiple parameters to be optimized and the computational complexity of performing non-linear analysis make it difficult to find the optimal design parameters in the realistic 3D structure. In this study, genetic algorithm-based optimal seismic retrofit methods for determining the required number, yield strength, and location of steel slit dampers are proposed to retrofit an asymmetric soft first-story structure. These methods use a multi-objective and single-objective evolutionary algorithms, each of which varies in computational complexity and incorporates nonlinear time-history analysis to determine seismic performance. Pareto-optimal solutions of the multi-objective optimization are found using a non-dominated sorting genetic algorithm (NSGA-II). It is demonstrated that the developed multi-objective optimization methods can determine the optimum number, yield strength, and location of dampers that satisfy the given limit state of a three-dimensional asymmetric soft first-story structure. It is also shown that the single-objective distribution method based on minimizing plan-wise stiffness eccentricity turns out to produce similar number of dampers in optimum locations without time consuming nonlinear dynamic analysis.

• Journal of the Society of Naval Architects of Korea
• /
• v.31 no.2
• /
• pp.91-98
• /
• 1994

An efficient optimization procedure combining the frequency approximation technique and the component-mode synthesis method is proposed for the structural dynamic optimization of the large structures subject to prescribed natural frequency constraints. Frequency constraints are approximated by using the first-order sensitivities with respect to both design parameters and their reciprocals. The component-mode synthesis method proposed by the authors in Ref.[8] is used for the repetitive detail finite-element analysis and sensitivity analysis. The validity of the proposed optimization procedure is confirmed through the numerical implementation of some examples. The presented approximation technique requires much smaller number of repetitive analysis than that using the sensitivities with respect to design parameters only, and further improvement in the numerical efficiency is achieved by the adoption of the introduced component-mode synthesis.

• Structural Engineering and Mechanics
• /
• v.32 no.5
• /
• pp.667-683
• /
• 2009

FE model-based dynamic analysis has been widely used to predict the dynamic characteristics of civil structures. In a physical point of view, an FE model is unavoidably different from the actual structure as being formulated based on extremely idealized engineering drawings and design data. The conventional model updating methods such as direct method and sensitivity-based parameter estimation are not flexible for model updating of complex and large structures. Thus, it is needed to develop a model updating method applicable to complex structures without restriction. The main objective of this paper is to present the model updating method based on the hybrid genetic algorithm (HGA) by combining the genetic algorithm as global optimization method and modified Nelder-Mead's Simplex method as local optimization method. This FE model updating method using HGA does not need the derivation of derivative function related to parameters and without application of complicated inverse analysis methods. In order to allow its application on diversified and complex structures, a commercial FEA tool is adopted to exploit previously developed element library and analysis algorithms. Moreover, an output-level objective function making use of measurement and analytical results is also presented to update simultaneously the stiffness and mass of the analysis model. The numerical examples demonstrated that the proposed method based on HGA is effective for the updating of the FE model of bridge structures.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.6
• /
• pp.2493-2503
• /
• 2018

Passive dynamic walking exhibits humanoid and energy efficient gaits. However, optimal design of passive walker at multi-variable level is not well studied yet. This paper presents a Chaotic Particle Swarm Optimization (CPSO) algorithm and applies it to the optimal design of flexible passive walker. Hip torsional stiffness and damping were incorporated into flexible biped walker, to imitate passive elastic mechanisms utilized in human locomotion. Hybrid dynamics were developed to model passive walking, and period-one gait was gained. The parameters global searching scopes were gained after investigating the influences of structural parameters on passive gait. CPSO were utilized to optimize the flexible passive walker. To improve the performance of PSO, multi-scroll Jerk chaotic system was used to generate pseudorandom sequences, and chaotic disturbance would be triggered if the swarm is trapped into local optimum. The effectiveness of CPSO is verified by comparisons with standard PSO and two typical chaotic PSO methods. Numerical simulations show that better fitness value of optimal design could be gained by CPSO presented. The proposed CPSO would be useful to design biped robot prototype.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.7
• /
• pp.821-826
• /
• 2011

CAE-based structural optimization techniques are applied for the design of a lightweight crane. The boom of the crane is designed by shape optimization with the shape of the cross section of the boom as the design variable. The design objective is mass minimization, and the static strength and dynamic stiffness of the system are set as the design constraints. Hyperworks, a commercial analysis and optimization software, is used for shape and topology optimization. In order to consistently change the shape of the elements of the boom with respect to the change in the shape of its cross section, the morphing function in Hyperworks is used. The support of the boom of the original model is simplified to model the design domain for topology optimization, which is discretized by using three-dimensional solid elements. The final result after shape and topology optimization is 19% and 17% reduction in the masses of the boom and support, respectively, without a deterioration in the system stiffness.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.11
• /
• pp.2824-2838
• /
• 2013

There has been a lot of attention paid recently to analyze dynamic human gestures that vary over time. Most attention to dynamic gestures concerns with spatio-temporal features, as compared to analyzing each frame of gestures separately. For accurate dynamic gesture recognition, motion feature extraction algorithms need to find representative features that uniquely identify time-varying gestures. This paper proposes a new feature-extraction algorithm using temporal self-similarity based on a hierarchical human model. Because a conventional temporal self-similarity method computes a whole movement among the continuous frames, the conventional temporal self-similarity method cannot recognize different gestures with the same amount of movement. The proposed model-based temporal self-similarity method groups body parts of a hierarchical model into several sets and calculates movements for each set. While recognition results can depend on how the sets are made, the best way to find optimal sets is to separate frequently used body parts from less-used body parts. Then, we apply a multiclass support vector machine whose optimization algorithm is based on structural support vector machines. In this paper, the effectiveness of the proposed feature extraction algorithm is demonstrated in an application for taebo gesture recognition. We show that the model-based temporal self-similarity method can overcome the shortcomings of the conventional temporal self-similarity method and the recognition results of the model-based method are superior to that of the conventional method.