• 품질경영학회지
• /
• 제29권3호
• /
• pp.86-96
• /
• 2001

The objective of this research is to develop a robust design methodology for plastic injection molded parts wherein warpage will be minimized by a complex method which is a kind of a simple direct search method. The design space considered for optimization is divided Into two sub-design space : mold and process conditions. Warpage is quantified using the Moldflow injection molding simulation software. The design methodology was applied to an actual part of a fax machine, the Guide-ASF model, through two different design policies. The significance of this study is the synthesis of a computer simulation of injection molding process and optimization technique to determine the optimal robust design solution.

• 한국기계가공학회지
• /
• 제7권1호
• /
• pp.53-61
• /
• 2008

In this research, the layout and geometry of U-type ribs in the part, including significant process conditions, are automatically optimized to reduce part warpage with robustness in consideration. The optimization procedure are based on an iterative redesign methodology integrated with computer aided injection molding simulation, Taguchi's Design of Experiment(DOE), and a direct search-based optimization method. The robustness of a design alternative is efficiently measured by introducing composite noise factor and Taguchi's signal-to-noise ratio. As a solution search methodology, the modified design space reduction method based on orthogonal arrays is employed to exploit an optimal robust design alternative. To illustrate the proposed methodology, a case study is performed on simulation results, where an optimal robust design alternative is obtained with a moderate number of iterations.

• 한국공간구조학회논문집
• /
• 제3권4호
• /
• pp.59-67
• /
• 2003

Genetic algorithm is one of the best ways to solve a discrete variable optimization problem. Genetic algorithm tends to thrive in an environment in which the search space is uneven and has many hills and valleys. In this study, genetic algorithm is used for solving the design problem of gable structure. The design problem of frame structure has some special features(complicate design space, many nonlinear constrants, integer design variables, termination conditions, special information for frame members, etc.), and these features must be considered in the formulation of optimization problem and the application of genetic algorithm. So, 'FRAME operator', a new genetic operator for solving the frame optimization problem effectively, is developed and applied to the design problem of gable structure. This example shows that the new opreator has the possibility to be an effective frame design operator and genetic algorithm is suitable for the frame optimization problem.

• 대한조선학회논문집
• /
• 제31권3호
• /
• pp.12-18
• /
• 1994

공학설계에 있어서 최적해를 얻기 위한 방법중의 하나로 최적화방법이 많이 사용되어 왔으나, 기존의 최적화방법에서는 설계점이 국부 최적점으로 빠져 들어갈 경우 그 영역을 벗어날 수 있는 방법이 없기 때문에, 최적화의 초기점을 달리하여 반복계산을 수행하여야 하는 불편한 점이 있었다. 유전적 알고리즘은 기존의 최적화방법에 비하여 다수의 설계점을 동시에 탐색하는 특성이 있어 국부 최적점에 빠질 가능성이 적은 반면, 계산시간이 많이 소요되고 전체 최적점 근처까지는 잘 수렴하나 정확한 최적점을 잘 찾지 못하는 한계가 있다. 본 연구에서는 유전적 알고리즘과 직접탐색법을 결합하여 이들의 단점을 보완한 즉, 전체 최적점을 보다 효율적으로 찾고 계산시간을 줄일 수 있는 방법을 제시하였다. 이 방법은 유전적 알고리즘을 이용하여 최적점이 존재하는 영역을 찾은 후에, 그 영역에서 직접탐색법을 이용하여 보다 정확한 최적점을 찾는 것으로, 예제를 통하여 제안된 방법의 유용성을 보였다.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제12권1호
• /
• pp.596-604
• /
• 2020

Previous studies on pipe auto-routing algorithms generally used such algorithms as A*, Dijkstra, Genetic Algorithm, Particle Swarm Optimization, and Ant Colony Optimization, to satisfy the relevant constraints of its own field and improve the output quality. On the other hand, this study aimed to significantly improve path-finding speed by applying the Jump Point Search (JPS) algorithm, which requires lower search cost than the abovementioned algorithms, for pipe routing. The existing JPS, however, is limited to two-dimensional spaces and can only find the shortest path. Thus, it requires several improvements to be applied to pipe routing. Pipe routing is performed in a three-dimensional space, and the path of piping must be parallel to the axis to minimize its interference with other facilities. In addition, the number of elbows must be reduced to the maximum from an economic perspective, and preferred spaces in the path must also be included. The existing JPS was improved for the pipe routing problem such that it can consider the above-mentioned problem. The fast path-finding speed of the proposed algorithm was verified by comparing it with the conventional A* algorithm in terms of resolution.

• 한국소음진동공학회논문집
• /
• 제15권11호
• /
• pp.1223-1231
• /
• 2005

Recently, the issue of ship nitration due to the large scale, high speed and lightweight of ship is emerging. For pleasantness in the cabin, shipbuilders are asked for strict vibration criteria and the degree of nitration level at a deckhouse became an important condition for taking order from customers. This study proposes a new optimization framework that is NASTRAN external call type optimization method (OptShip) and applies to an optimum design to decrease the nitration level of a deckhouse. The merits of this method are capable of using of global searching method and selecting of various objective function and design variables. The global optimization algorithms used here are random tabu search method which has fast converging speed and searches various size domains and genetic algorithm which searches multi-point solutions and has a good search capability in a complex space. By adapting OptShip to full-scale model, the validity of the suggested method was investigated.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 추계학술대회 논문집 학회본부
• /
• pp.531-533
• /
• 1997

PI(proportional-integral) controller has been extensively used in the industrial field. But in practicle case, it is difficult to tune PI gains. Evolution Strategy(ES) is used as an effective search algorithm in optimization programs. In this paper we proposed a PI controller for Spot welder system using ES with varying search space. ES with varying search space which depends on fitness values at each generation is used to tune PI control parameters. Simulation results show the proposed algorithm has accurate and robust performance with effective search ability.

• Journal of Electrical Engineering and Technology
• /
• 제7권4호
• /
• pp.636-645
• /
• 2012

This study introduces an information granular-based fuzzy radial basis function neural networks (FRBFNN) based on multiobjective optimization and weighted least square (WLS). An improved multiobjective space search algorithm (IMSSA) is proposed to optimize the FRBFNN. In the design of FRBFNN, the premise part of the rules is constructed with the aid of Fuzzy C-Means (FCM) clustering while the consequent part of the fuzzy rules is developed by using four types of polynomials, namely constant, linear, quadratic, and modified quadratic. Information granulation realized with C-Means clustering helps determine the initial values of the apex parameters of the membership function of the fuzzy neural network. To enhance the flexibility of neural network, we use the WLS learning to estimate the coefficients of the polynomials. In comparison with ordinary least square commonly used in the design of fuzzy radial basis function neural networks, WLS could come with a different type of the local model in each rule when dealing with the FRBFNN. Since the performance of the FRBFNN model is directly affected by some parameters such as e.g., the fuzzification coefficient used in the FCM, the number of rules and the orders of the polynomials present in the consequent parts of the rules, we carry out both structural as well as parametric optimization of the network. The proposed IMSSA that aims at the simultaneous minimization of complexity and the maximization of accuracy is exploited here to optimize the parameters of the model. Experimental results illustrate that the proposed neural network leads to better performance in comparison with some existing neurofuzzy models encountered in the literature.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2002년도 추계종합학술대회
• /
• pp.251-254
• /
• 2002

이동 무선 통신 시스템을 설계할 때 기지국(BTS), 기지국 콘트롤러(BSC), 이동 교환국(MSC)의 위치는 매우 중요한 파라미터 중 하나이다. 무선 통신 시스템을 설계할 때 여러 가지 복잡한 변수들을 잘 조합하여 비용이 최소가 되도록 설계해야 한다. 이러한 문제를 해결하는데 필요한 알고리즘이 전역 최적화 알고리즘이며, 지금까지 전역 최적화 검색 기술로 Random Walk, Simulated Annealing, Tabu Search, Genetic Algorithm과 같은 조합 최적화 기술이 사용되어 왔다. 본 논문은 이동 통신 시스템의 기지국, 기지국 콘트롤러, 이동 교환국의 위치 최적화에 위의 4가지 알고리듬들을 적용하여 각 알고리듬의 결과를 비교 분석하며 알고리듬에 의한 최적화 과정을 보여 준다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2000년도 춘계학술대회 학술발표 논문집
• /
• pp.241-244
• /
• 2000

The optimization method based on an enhanced genetic algorithms is proposed for multimodal function optimization in this paper This method is consisted of two main steps. The first step is global search step using the genetic algorithm(GA) and function assurance criterion(FAC). The belonging of an population to initial solution group is decided according to the FAC. The second step is to decide resemblance between individuals and research optimum solutions by single point method in reconstructive research space. Two numerical examples are also presented in this paper to comparing with conventional methods.