• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.2331-2336
• /
• 2005

This paper proposes a new haptic device using a parallel mechanism with gimbal type actuators. This device has three legs actuated by 2-DOF gimbal mechanisms, which make the device simple and light by fixing all the actuators to the base. Three extra sensors are placed at passive joints to obtain a unique solution of the forward kinematics problem. The proposed haptic device is developed for an operator to use it on a desktop in due consideration of the size of an average Korean. The proposed haptic device has a small workspace for on operator to use it on a desktop and more sensitivity than a serial type haptic device. Therefore, the motors of the proposed haptic device are fixed at the base plate so that the proposed haptic device has a better dynamic bandwidth due to a low moving inertia. With this conceptual design, optimization of the design parameters is carried out. The objective function is defined by the fuzzy minimum of the global design indices, global force/moment isotropy index, global force/moment payload index, and workspace. Each global index is calculated by a SVD (singular value decomposition) of the force and moment parts of the jacobian matrix. Division of the jacobian matrix assures a consistency of the units in the matrix. Due to the nonlinearity of this objective function, Genetic algorithms are adopted for a global optimization.

• 전력전자학회논문지
• /
• 제27권4호
• /
• pp.297-304
• /
• 2022

This study compares the performance of artificial intelligence (AI)-based maximum power point tracking (MPPT) methods under partial shading conditions in a photovoltaic generation system. Although many studies on AI-based MPPT have been conducted, few studies comparing the tracking performance of various AI-based global MPPT methods seem to exist in the literature. Therefore, this study compares four representative AI-based global MPPT methods including fuzzy logic control (FLC), particle swarm optimization (PSO), grey wolf optimization (GWO), and genetic algorithm (GA). Each method is theoretically analyzed in detail and compared through simulation studies with MATLAB/Simulink under the same conditions. Based on the results of performance comparison, PSO, GWO, and GA successfully tracked the global maximum power point. In particular, the tracking speed of GA was the fastest among the investigated methods under the given conditions.

• 한국시뮬레이션학회논문지
• /
• 제25권1호
• /
• pp.1-9
• /
• 2016

본 연구는 공대공 미사일 조종날개의 공력 및 구조를 동시에 고려한 구동력 최소화에 대한 최적화를 수행하였다. 본 연구에서는 조종날개의 공력 및 구조적 특성을 동시에 고려하기 위하여 공력-구조 연계 시뮬레이션을 사용하였으며 공력 및 구조 시뮬레이션에 각각의 전용 소프트웨어를 사용하고자 비정상-약결합 방식 연계기법을 적용하였다. 전역 최적화에는 많은 반복 계산이 필요하므로 빠른 계산을 위하여 수학적 모델링을 이용하였으며 이를 위하여 면 중앙 합성 실험계획법으로 실험점을 선정하였다. 선정된 실험점 및 그에 대한 공력-구조 연계 시뮬레이션 결과를 토대로 2차 다항식 반응면을 생성하였으며 생성된 수학적 모델링을 이용, 유전자 알고리즘 기반 전역최적 설계를 수행하였다. 최적화 목적함수는 마하 0.7 및 마하 2.0 사이의 압력 중심점 이동거리 최소화로 설정하였으며 최적화 결과 압력 중심점 이동거리가 7.5% 감소된 최적형상을 도출하였다.

• 한국생산제조학회지
• /
• 제23권3호
• /
• pp.213-217
• /
• 2014

A hybrid search method based on the artificial bee colony algorithm (ABCA) with harmony search (HS) is suggested for finding a global solution in the field of optimization. Three cases of the suggested algorithm were examined for improving the accuracy and convergence rate. The results showed that the case in which the harmony search was implemented with the onlooker phase in ABCA was the best among the three cases. Although the total computation time of the best case is a little bit longer than the original ABCA under the prescribed conditions, the global solution improved and the convergence rate was slightly faster than those of the ABCA. It is concluded that the suggested algorithm improves the accuracy and convergence rate, and it is expected that it can effectively be applied to optimization problems with many design variables and local solutions.

• 한국안광학회지
• /
• 제6권2호
• /
• pp.121-127
• /
• 2001

additive 감쇠에 의한 감쇠 최소 자승법에 가우스 소거법과 Jacobian 행렬을 직교 변환시킨 SVD(singular value decomposition)법을 적용하여 조건수가 양호한 triplet 사진 렌즈계에 적용하여 수렴 속도와 안정성을 비교하였다. SVD 직교화 방법을 적용한 감쇠 최소 자승 법이 최소 merit 함수에 보다 안정되고 빠르게 수렴하였다. SVD 방법을 적용한 최적화에서 적절한 merit 함수를 얻을 수 있지만 오차 함수의 비선형성으로 인해 merit 함수가 국부 최소 점에 수업하는 경우가 있어서 간단한 전역 최적화 방법인격자 법으로 최적화를 실시하여 SVD 방법에 의한 merit 함수보다 낮은 전역 최소 점에 수렴하게 하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.324-329
• /
• 2005

Dynamic Encoding Algorithm for Searches (DEAS) which is known as a fast and reliable non-gradient optimization method, was proposed [1]. DEAS reaches local or global optimum with binary strings (or binary matrices for multi-dimensional problem) by iterating the two operations; bisectional search (BSS) and unidirectional search (UDS). BSS increases binary strings by one digit (i.e., 0 or 1), while UDS performs increment or decrement of binary strings in the BSS' result direction with no change of string length. Because the interval of UDS exponentially decreases with increment of bit string length (BSL), DEAS is difficult to escape from local optimum when DEAS falls into local optimum. Therefore, this paper proposes hopping UDS (HUDS) which performs UDS by hopping as many as BSL in the final point of UDS process. HUDS helps to escape from local optimum and enhances a probability searching global optimization. The excellent performance of HUDS will be validated through the well-known benchmark functions.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1283-1286
• /
• 2005

Multimodal optimization is one of the most interesting topics in evolutionary computational discipline. Simple genetic algorithm, a basic and good-performance genetic algorithm, shows bad performance on multimodal problems, taking long generation time to obtain the optimum, converging on the local extrema in early generation. In this paper, we propose a new genetic algorithm with two new genetic mutational operators, i.e. global and local mutation operators, and no genetic crossover. The proposed algorithm is similar to Simple GA and the two genetic operators are as simple as the conventional mutation. They just mutate the genes from left or right end of a chromosome till the randomly selected gene is replaced. In fact, two operators are identical with each other except for the direction where they are applied. Their roles of shaking the population (global searching) and fine tuning (local searching) make the diversity of the individuals being maintained through the entire generation. The proposed algorithm is, therefore, robust and powerful.

• 전자공학회논문지B
• /
• 제31B권4호
• /
• pp.145-154
• /
• 1994

This paper proposes an efficient method for improving the training performance of the neural network by using a hybrid of a stochastic approximation and a backpropagation algorithm. The proposed method improves the performance of the training by appliying a global optimization method which is a hybrid of a stochastic approximation and a backpropagation algorithm. The approximate initial point for a stochastic approximation and a backpropagation algorihtm. The approximate initial point for fast global optimization is estimated first by applying the stochastic approximation, and then the backpropagation algorithm, which is the fast gradient descent method, is applied for a high speed global optimization. And further speed-up of training is made possible by adjusting the training parameters of each of the output and the hidden layer adaptively to the standard deviation of the neuron output of each layer. The proposed method has been applied to the parity checking and the pattern classification, and the simulation results show that the performance of the proposed method is superior to that of the backpropagation, the Baba's MROM, and the Sun's method with randomized initial point settings. The results of adaptive adjusting of the training parameters show that the proposed method further improves the convergence speed about 20% in training.

• 대한기계학회논문집A
• /
• 제29권10호
• /
• pp.1369-1376
• /
• 2005

In structural design, the design variables are frequently selected from certain discrete values. Various optimization algorithms have been developed fDr discrete design. It is well known that many function evaluations are needed in such optimization. Recently, sequential algorithm with orthogonal arrays (SOA), which is a search algorithm for a local minimum in a discrete space, has been developed. It considerably reduces the number of function evaluations. However, it only finds a local minimum and the final solution depends on the initial values of the design variables. A new algorithm is proposed to adopt a genetic algorithm (GA) in SOA. The GA can find a solution in a global sense. The solution from the GA is used as the initial design of SOA. A sequential usage of the GA and SOA is carried out in an iterative manner until the convergence criteria are satisfied. The performance of the algorithm is evaluated by various examples.

• 한국산학기술학회논문지
• /
• 제21권2호
• /
• pp.8-18
• /
• 2020

본 연구에서는 기존의 Harmony Search(HS)의 성능을 강화한 Exponential Bandwidth Harmony Search with Centralized Global Search(EBHS-CGS)를 개발하였다. EBHS-CGS는 HS의 성능 강화를 위해 총 두 가지 방법을 추가하였다. 첫 번째 방법은 지역탐색을 강화하기 위한 Bandwidth(bw) 개량방안이다. 이 방법은 기존 bw를 지수형태의 bw로 대체하여 적용함으로써 반복시산이 진행되면서 bw값을 줄인다. 이러한 형태의 bw는 정밀한 지역탐색을 가능하고, 이를 통해 알고리즘은 더욱 정밀한 값을 구할 수 있다. 두 번째 방법은 효과적인 전역탐색을 위한 탐색범위 축소이다. 이 방법은 Harmony Memory(HM) 내에서 가장 좋은 결정변수를 고려하여 탐색범위를 축소한다. 이를 Centralized Global Search(CGS)라 하며, 이 과정은 새로운 매개변수 Centralized Global Search Rate(CGSR)에 의해 HS의 전역탐색과는 별도로 진행된다. 축소된 탐색범위는 효과적인 전역탐색을 가능하게 하며, 이를 통해 알고리즘의 성능이 향상된다. EBHS-CGS를 대표적인 최적화 문제(수학 및 공학 분야)에 적용하고, 그 결과를 HS와 Improved Harmony Search(IHS)와 비교하여 제시하였다.