• 제어로봇시스템학회논문지
• /
• 제13권4호
• /
• pp.315-319
• /
• 2007

A technique to model plasma processes was presented. This was accomplished by combining the backpropagation neural network (BPNN) and genetic algorithm (GA). Particularly, the GA was used to optimize five training factor effects by balancing the training and test errors. The technique was evaluated with the plasma etch data, characterized by a face-centered Box Wilson experiment. The etch outputs modeled include Al etch rate, AI selectivity, DC bias, and silica profile angle. Scanning electron microscope was used to quantify the etch outputs. For comparison, the etch outputs were modeled in a conventional fashion. GABPNN models demonstrated a considerable improvement of more than 25% for all etch outputs only but he DC bias. About 40% improvements were even achieved for the profile angle and AI etch rate. The improvements demonstrate that the presented technique is effective to improving BPNN prediction performance.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.1059-1060
• /
• 2008

In this paper, using a genetic algorithm, consisting of six to seven degrees of freedom links, walking robot to up-stair that can walk to optimize energy and stability to generate. Walking robot to up-stairs of the four-step segmentation of the various situations that match the pace and pattern so that it can generate. It also generated using genetic algorithms to test for Matlab into the Robot Simulation of the humanoid experiment was used.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 추계학술대회 논문집 학회본부
• /
• pp.174-176
• /
• 1997

This paper proposes a unit commitment scheduling method based on hybrid genetic algorithm(GA). When the systems are scaled up, conventional genetic algorithms suffer from computational time limitations because of the growth of the search space. So greatly reduce the search space of the GA and to efficiently deal with the constraints of the problem, priority list unit ordering scheme are incorporated as the initial solution and the minimum up and down time constraints of the units are included. The violations of other constraints are handled by integrating penalty factors. To show the effectiveness of the proposed method. test results for system of 10 units is compared with results obtained using other methods.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1995년도 추계학술대회 논문집 학회본부
• /
• pp.64-66
• /
• 1995

This paper presents a new approach to evaluate optimal basic reliability indices of electric distribution systems using genetic algorithm. The use of optimal reliability evaluation is an important aspect of distribution system planning and operation to determine adequacy reliability level of each area. In this paper, the reliability model is based on the analytical method, connecting component failure to load point outage in each section. The proposed method applies genetic algorithm to calculate the optimal values of basic reliability indices, ie. failure rate and repair time, for a load point in the power distribution system, subject to minimizing interruption cost. Test results for the model system are reported in the paper compared with a direct optimization method(gradient projection).

• 한국국방경영분석학회지
• /
• 제24권2호
• /
• pp.117-129
• /
• 1998

This study is concerned with developing a heuristic for a web-based vehicle routing problem using mixed genetic algorithm(VRPMGA) which determines each vehicle route in order to minimize the transportation costs, subject to meeting the demands of all delivery points. VRP is known to be NP-hard, and it needs a lot of computing time to get the optimal solution, so that heuristics are more frequently developed than optimal algorithms. This study aims to develop a mixed genetic algorithm by partitioned strategy which can give a good solution in comparatively brief time. The good features of the VRPMGA are, fristly, the ability of early convergence and, secondly, the capability of producing multiple, alternative, and near-optimal solutions. The VRPMGA is a useful algorithm that can be appliable to VRP and TSP. Finally, the computational test were performed using the benchmark problems and the proposed heuristic is compared with the other existing algorithms (COSA). The result of computational tests shows that proposed heuristic gives good solutions, in much shorter time, which are same as the best known solutions in the pervious research.

• 대한산업공학회지
• /
• 제26권2호
• /
• pp.183-192
• /
• 2000

This paper considers the problem of scheduling a set of n jobs on m parallel machines to minimize total weighted tardiness. For the problem a genetic algorithm is proposed, in which solutions are encoded using the random key method suggested by Bean and new crossover operators are employed to increase performance of the algorithm. The algorithm is compared with the Modified Due-Date (MDD) algorithm after series of tests to find appropriate values for genetic parameters. Results of computational tests on randomly generated test problems show that the suggested algorithm performs better than the MDD algorithm and gives good solutions in a reasonable amount of computation time.

• 한국지능시스템학회논문지
• /
• 제21권2호
• /
• pp.206-211
• /
• 2011

본 논문에서는 연속적인 파라미터 공간에 대한 최적화에 대해 진화적 알고리즘의 특징적인 형상화를 제시한다. 이 방법은유전알고리즘이 연속적인 탐색공간에서의 파라미터 식별에 대해 가장 강점을 지녔다는 점에 착안한 것이다. 유전알고리즘과 제안한 알고리즘과의 주요한 차이점은 개별적 또는 연속적인 묘사의 차이가 있다는 것이다. 잘 알려진 실험함수의 최적화문제를 도입하여 연속 탐색공간 문제에 대해 제안하는 알고리즘에 대해 계산시간 및 사용메모리 등의 성능이 우수하다는 효율성을 보였다.

• 대한기계학회논문집A
• /
• 제27권4호
• /
• pp.499-506
• /
• 2003

The design cycle associated with large engineering systems requires an initial decomposition of the complex system into design processes which are coupled through the transference of output data. Some of these design processes may be grouped into iterative subcycles. In analyzing or optimizing such a coupled system, it is essential to determine the best order of the processes within these subcycles to reduce design cycle time and cost. This is accomplished by decomposing large multidisciplinary problems into several sub design structure matrices (DSMs) and processing them in parallel This paper proposes a new method for parallel decomposition of multidisciplinary problems to improve design efficiency by using the multi-objective genetic algorithm and two sample test cases are presented to show the effect of the suggested decomposition method.

• 한국작물학회지
• /
• 제46권1호
• /
• pp.12-16
• /
• 2001

In order to develop near isogenic lines (NILs) the heterozygous rice plants for alkali digestibility value(ADV) were selected and tested in every generation from $F_5$to $F_9$ of a cross, Seratus malam/Suwon 345. Finally several sets of NILs, which were six low ADV lines and four medium-high ADV lines, were selected among $F_{10}$ lines. No differences of the plant growth characteristics, amylose content and protein content of rice grain were found between low and medium-high ADV lines. Rice flour of low ADV lines showed longer gel length in gel consistency test than medium-high ADV lines, and also showed different gelatinization characteristics measured by rapid visco analyzer compared with medium-high ADV lines. The result of genetic analysis indicated that ADV-NILs developed were differed in a pair of major gene controlling ADV, and low ADV was dominant over medium-high ADV.

• EDISON SW 활용 경진대회 논문집
• /
• 제4회(2015년)
• /
• pp.476-478
• /
• 2015

This paper deals with synthesis of 4 bar linkage by using optimum design. To design 4 bar linkage, overlay method is proposed and for optimization, genetic algorithm is applied with objective function. The accuracy of this method will be determined by errors between real value and test value. We will use Chebychev spacing to get 3 precision positions of input angles. The output angles will be determined by the function that the designer wants input and output relations to be. It will be applied to example to show the accuracy of this method. The advantages of using this method are that it is fast to get optimal solution and it is simple to use.