• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 추계학술대회 논문집 학회본부
• /
• pp.6-8
• /
• 1996

The multiobjective optimization is presented for the optimal design of induction motors. The aim of design is to find an optimized induction motor in terms of both the efficiency and the mass. The efficiency and the mass are linearly combined using the weighting factors. Optimization process is performed by using the improved (1+1) evolution strategy (ES). ES is the algorithm that can find the global minimum. To verify the validity of the proposed method. the method is applied to a sample design.

• 한국항만학회지
• /
• 제12권2호
• /
• pp.217-224
• /
• 1998

During the operation of crane system in container yard, the objective is to transport the load to a goal position as quick as possible without rope oscillation. The container crane is generally operated by an expert operator, but recently an automatic control system with high speed and rapid transportation is required. Therefore, we developed an optimal controller which has to control the crane system with disturbances. In this paper, we present a design of optima 2-DOF PID controller for the control of gantry crane which has to control swing motion and trolley position. We used evolution strategy(ES) to tune the parameters of 2-DOF PID controller. It was compared with general PID controller. The computer simulations show that the proposed method has better performances than the other method.

• 경영과학
• /
• 제28권3호
• /
• pp.47-60
• /
• 2011

While the literature is replete with arguments that corporate strategy, structure, and culture independently matter in explaining the growth and survival of firms, little theoretical and empirical attention has been devoted to understanding how the three organizational factors develop over time in interaction. Through in-depth case study, we examine the POSCO's historical development with respect to the strategy-structure-culture interface. Furthermore, by looking at the POSCO's longitudinal financial data we gain insight into whether the aligned evolution of strategy, structure, and culture is associated with performance.

• Smart Structures and Systems
• /
• 제15권3호
• /
• pp.735-749
• /
• 2015

The major difficulty of using Bayesian probabilistic inference for system identification is to obtain the posterior probability density of parameters conditioned by the measured response. The posterior density of structural parameters indicates how plausible each model is when considering the uncertainty of prediction errors. The Markov chain Monte Carlo (MCMC) method is a widespread medium for posterior inference but its convergence is often slow. The differential evolution adaptive Metropolis-Hasting (DREAM) algorithm boasts a population-based mechanism, which nms multiple different Markov chains simultaneously, and a global optimum exploration ability. This paper proposes an improved differential evolution adaptive Metropolis-Hasting algorithm (IDREAM) strategy to estimate the posterior density of structural parameters. The main benefit of IDREAM is its efficient MCMC simulation through its use of the adaptive Metropolis (AM) method with a mutation strategy for ensuring quick convergence and robust solutions. Its effectiveness was demonstrated in simulations on identifying the structural parameters with limited output data and noise polluted measurements.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.173-176
• /
• 2001

This paper proposes training algorithms of neuro-fuzzy systems. First, we introduce a structure training algorithm, which produces the necessary number of hidden nodes from training data. From this algorithm, initial fuzzy rules are also obtained. Second, the parameter training algorithm using evolution strategy is introduced. In order to show their usefulness, we apply our neuro-fuzzy system to a nonlinear system identification problem. It was found from experiments that proposed training algorithms works well.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 춘계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.106-108
• /
• 2006

This paper shows the design of a levitation magnet for an OLED system which has a large air gap. Evolution strategy was adopted for the optimization of the magnet system. During the optimization process, interpolation of levitation force was used to reduce the computation time which was needed to calculate the levitation force. Object function for optimization was total weight of the magnet system.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 A
• /
• pp.30-33
• /
• 1997

이 논문에서는 코깅 토크의 최소화와 효율의 최대화를 위해 전동기의 영구자석과 고정자 슬롯의 형상 최적화 방법을 제시하였다. 자계 해석은 유한요소법을 이용하였고, 토크의 계산은 가상번위법에 의하여 수행하였다. 형상 최적화를 위해서는 다중 목적 프로그래밍 기법과 개선된 ES를 적용하였다. 그리고 결과로 최적 설계된 전동기를 초기 전동기와 비교하였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제1권1호
• /
• pp.125-130
• /
• 2001

In this paper, we propose a new type of evolution strategies combined with reinforcement learning. We use the variances of fitness occurred by mutation to make the reinforcement signals which estimate and control the step length of mutation. With this proposed method, the convergence rate is improved. Also, we use cauchy distributed mutation to increase global convergence faculty. Cauchy distributed mutation is more likely to escape from a local minimum or move away from a plateau. After an outline of the history of evolution strategies, it is explained how evolution strategies can be combined with the reinforcement learning, named reinforcement evolution strategies. The performance of proposed method will be estimated by comparison with conventional evolution strategies on several test problems.

• 센서학회지
• /
• 제14권6호
• /
• pp.402-408
• /
• 2005

This paper describes the optimal design of a multi-layered surface plasmon resonance sensors to meet various specifications and improve some physical parameters. Dip 3 dB bandwidth and depth were chosen as design parameters and the objective function was the norm of the difference between design parameters and target values. The design variables are thicknesses of each layer and to obtain the design parameters, the optical admittance method was employed. The (1+1) evolution strategy was employed as an optimization tool. By applying the proposed optimization procedure to a 3-layered sensor, the optimized design variables considerably improved the 3 dB bandwidth by 4.8 nm and the dip depth by 1.1 dB.

• Advances in Computational Design
• /
• 제1권3호
• /
• pp.275-296
• /
• 2016

This article presents an adaptive directional differential evolution (ADDE) algorithm and its application in solving discrete sizing truss optimization problems. The algorithm is featured by a new self-adaptation approach and a simple directional strategy. In the adaptation approach, the mutation operator is adjusted in accordance with the change of population diversity, which can well balance between global exploration and local exploitation as well as locate the promising solutions. The directional strategy is based on the order relation between two difference solutions chosen for mutation and can bias the search direction for increasing the possibility of finding improved solutions. In addition, a new scaling factor is introduced as a vector of uniform random variables to maintain the diversity without crossover operation. Numerical results show that the optimal solutions of ADDE are as good as or better than those from some modern metaheuristics in the literature, while ADDE often uses fewer structural analyses.