• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.606-609
• /
• 2015

본 연구에서는 support vector regression (SVR) 및 매개변수 최적화 알고리즘을 이용한 하천수위 예측모델을 구축하고 이를 실제 유역에 적용하여 모델 효율성을 평가하였다. 여기서, SVR은 하천수위를 예측하기 위한 예측모델로서 채택되었으며, 커널함수 (Kernel function)로서는 radial basis function (RBF)을 선택하였다. 최적화 알고리즘은 SVR의 최적 매개변수 (C?, cost parameter or regularization parameter; ${\gamma}$, RBF parameter; ${\epsilon}$, insensitive loss function parameter)를 탐색하기 위하여 적용되었다. 매개변수 최적화 알고리즘으로는 grid search (GS), genetic algorithm (GA), particle swarm optimization (PSO), artificial bee colony (ABC) 알고리즘을 채택하였으며, 비교분석을 통해 최적화 알고리즘의 적용성을 평가하였다. 또한 SVR과 최적화 알고리즘을 결합한 모델 (SVR-GS, SVR-GA, SVR-PSO, SVR-ABC)은 기존에 수자원 분야에서 널리 적용되어온 신경망(Artificial neural network, ANN) 및 뉴로퍼지 (Adaptive neuro-fuzzy inference system, ANFIS) 모델과 비교하였다. 그 결과, 모델 효율성 측면에서 SVR-GS, SVR-GA, SVR-PSO 및 SVR-ABC는 ANN보다 우수한 결과를 나타내었으며, ANFIS와는 비슷한 결과를 나타내었다. 또한 SVR-GA, SVR-PSO 및 SVR-ABC는 SVR-GS보다 상대적으로 우수한 결과를 나타내었으며, 모델 효율성 측면에서 SVR-PSO 및 SVR-ABC는 가장 우수한 모델 성능을 나타내었다. 따라서 본 연구에서 적용한 매개변수 최적화 알고리즘은 SVR의 매개변수를 최적화하는데 효과적임을 확인할 수 있었다. SVR과 최적화 알고리즘을 이용한 하천수위 예측모델은 기존의 ANN 및 ANFIS 모델과 더불어 하천수위 예측을 위한 효과적인 도구로 사용될 수 있을 것으로 판단된다.

• Journal of Korea Water Resources Association
• /
• v.34 no.5
• /
• pp.473-486
• /
• 2001

The WGR model was developed to represent meso-scale precipitation. As a conceptual model, this model shows a good link between atmospheric dynamics and statistical description of meso-scale precipitation(Waymire et al., 1984). However, as it has maximum 18 parameters along with its non-linear structure, its parameter estimation has been remained a difficult problem. There have been several cases of its parameter estimation for different fields using non-linear programming techniques(NLP), which were also difficult tasks to hamper its wide applications. In this study, we estimated the WGR model parameters of the Han river basin using the genetic algorithm(GA) and compared them to the NLP results(Yoo and Kwon, 2000). As a result of the study, we can find that the sum of square error from the GA provide more consistent parameters to the seasonal variation of rainfall. Also, we can find that the higher rainfall amount during summer season is closely related with the arrival rate of rain bands, not the rain cell intensity.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.3
• /
• pp.215-222
• /
• 2001

Neural network based models of semiconductor manufacturing processes have been shown to offer advantages in both accuracy and generalization over traditional methods. However, model development is often complicated by the fact that back-propagation neural networks contain several adjustable parameters whose optimal values unknown during training. These include learning rate, momentum, training tolerance, and the number of hidden layer neurOnS. This paper presents an investigation of the use of genetic algorithms (GAs) to determine the optimal neural network parameters for the modeling of plasma-enhanced chemical vapor deposition (PECVD) of silicon dioxide films. To find an optimal parameter set for the neural network PECVD models, a performance index was defined and used in the GA objective function. This index was designed to account for network prediction error as well as training error, with a higher emphasis on reducing prediction error. The results of the genetic search were compared with the results of a similar search using the simplex algorithm.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2001.05a
• /
• pp.341-346
• /
• 2001

• Transactions of the Korean Society of Automotive Engineers
• /
• v.12 no.3
• /
• pp.51-58
• /
• 2004

Optimization of engine desist and operation parameters using a genetic algorithm was demonstrated for direct injection diesel engine combustion. A micro genetic algorithm and a modified KIVA-3V code were used for the analysis and optimization of the engine combustion. At each generation of the optimization step the micro genetic algorithm generated five groups of parameter sets, and the five cases of KIVA-3V analysis were to be performed either in series or in parallel. The micro genetic algorithm code was also parallelized by using MPI programming, and a multi-CPU parallel supercomputer was used to speed up the optimization process by four times. An example case for a fixed engine speed was performed with six parameters of intake swirl ratio, compression ratio, fuel injection included angle, injector hole number, SOI, and injection duration. A simultaneous optimization technique for the whole range of engine speeds would be suggested for further studies.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.1639-1644
• /
• 2003

A hybrid genetic algorithm is developed for estimating the wall emissivities for an absorbing, emitting, and scattering media in a two-dimensional irregular geometry with diffusely emitting and reflecting opaque boundaries by minimizing an objective function, which is expressed by the sum of square errors between estimated and measured temperatures at only four data positions. The finite-volume method was employed to solve the radiative transfer equation for a two-dimensional irregular geometry. The results show that a developed hybrid genetic algorithms reduce the effect of genetic parameters on the performance of genetic algorithm and that the wall emissivities are estimated accurately without measurement errors.

• Proceedings of the Safety Management and Science Conference
• /
• 2002.05a
• /
• pp.91-97
• /
• 2002

In this thesis presents line balancing problems of two-sided and mixed model assembly line widely used in practical fields using genetic algorithm for reducing throughput time, cost of tools and fixtures and improving flexibility of assembly lines. Two-sided and mixed model assembly line is a special type of production line where variety of product similar in product characteristics are assembled in both sides. This thesis proposes the genetic algorithm adequate to each step in tow-sided and mixed model assembly line with suitable presentation, individual, evaluation function, selection and genetic parameter. To confirm proposed genetic algorithm, we apply to increase the number of tasks in case study. And for evaluation the performance of proposed genetic algorithm, we compare to existing algorithm of one-sided and mixed model assembly line. The results show that the algorithm is outstanding in the problems with a larger number of stations or larger number of tasks.

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.8
• /
• pp.1222-1226
• /
• 2002

Estimates of genetic correlation between direct and maternal effects for weaning weight of beef cattle are often negative in field data. The biological existence of this genetic antagonism has been the point at issue. Some researchers perceived such negative estimate to be an artifact from poor modeling. Recent studies on sources affecting the genetic correlation estimates are reviewed in this article. They focus on heterogeneity of the correlation by sex, selection bias caused from selective reporting, selection bias caused from splitting data by sex, sire by year interaction variance, and sire misidentification and inbreeding depression as factors contributing sire by year interaction variance. A biological justification of the genetic antagonism is also discussed. It is proposed to include the direct-maternal genetic covariance in the analytical models.

• Korean Journal of Computational Design and Engineering
• /
• v.14 no.6
• /
• pp.382-389
• /
• 2009

Recently, knowledge management has been required in companies as a tool of competitiveness. Companies have constructed Enterprise Resource Planning(ERP) system in order to manage huge knowledge. But, it is not easy to formalize knowledge in organization. We focused on data mining system by genetic programming(GP). Data mining system by genetic programming can be useful tools to derive and extract the necessary information and knowledge from the huge accumulated data. However when we don't have enough amounts of data to perform the learning process of genetic programming, we have to reduce input parameter(s) or increase number of learning or training data. In this study, an enhanced data mining method combining Genetic Programming with Self organizing map, that reduces the number of input parameters, is suggested. Experiment results through a prototype implementation are also discussed.

• Journal of the Korea Safety Management & Science
• /
• v.4 no.2
• /
• pp.83-101
• /
• 2002

In this thesis presents line balancing problems of two-sided and mixed model assembly line widely used in practical fields using genetic algorithm for reducing throughput time, cost of tools and fixtures and improving flexibility of assembly lines. Two-sided and mixed model assembly line is a special type of production line where variety of product similar in product characteristics are assembled in both sides. This thesis proposes the genetic algorithm adequate to each step in tow-sided and mixed model assembly line with suitable presentation, individual, evaluation function, selection and genetic parameter. To confirm proposed genetic algorithm, we apply to increase the number of tasks in case study. And for evaluation the performance of proposed genetic algorithm, we compare to existing algorithm of one-sided and mixed model assembly line. The results show that the algorithm is outstanding in the problems with a larger number of stations or larger number of tasks.