• 한국지능시스템학회논문지
• /
• 제18권4호
• /
• pp.530-536
• /
• 2008

무선 센서 네트워크에서 노드들의 위치 측정 문제는 다양한 위치인식 기법을 적용하기 위해 기본적으로 해결해야 한다. 위치측정 문제는 위치가 알려진 일부 고정 노드들을 기준으로 나머지 노드들의 위치를 결정하는 문제이다. 기존의 대부분의 위치측정 기법은 고정 노드들로부터의 각도나 거리 측정값을 기반으로 삼각기법(triangulation) 이나 multilateration 방법을 사용한다. 본 논문에서는 노드들 간의 연결성이 알려져 있을 때 퍼지 모델링과 유전자 알고리즘을 사용하여 서로 전송범위 내에 있는 노드 쌍에 대해 가중치를 주어 미지 노드의 위치를 계산하는 향상된 중점 기법을 제안한다. 또한 시뮬레이션을 통해 제안된 중점 기법이 단순히 연결성만을 사용하는 중점 기법에 비해 정확한 위치측정이 가능함을 확인하였다.

• Journal of Microbiology and Biotechnology
• /
• 제22권7호
• /
• pp.939-946
• /
• 2012

Antibiotic production with Streptomyces sindenensis MTCC 8122 was optimized under submerged fermentation conditions by artificial neural network (ANN) coupled with genetic algorithm (GA) and Nelder-Mead downhill simplex (NMDS). Feed forward back-propagation ANN was trained to establish the mathematical relationship among the medium components and length of incubation period for achieving maximum antibiotic yield. The optimization strategy involved growing the culture with varying concentrations of various medium components for different incubation periods. Under non-optimized condition, antibiotic production was found to be $95{\mu}g/ml$, which nearly doubled ($176{\mu}g/ml$) with the ANN-GA optimization. ANN-NMDS optimization was found to be more efficacious, and maximum antibiotic production ($197{\mu}g/ml$) was obtained by cultivating the cells with (g/l) fructose 2.7602, $MgSO_4$ 1.2369, $(NH_4)_2PO_4$ 0.2742, DL-threonine 3.069%, and soyabean meal 1.952%, for 9.8531 days of incubation, which was roughly 12% higher than the yield obtained by ANN coupled with GA under the same conditions.

• 한국자동차공학회논문집
• /
• 제23권6호
• /
• pp.565-574
• /
• 2015

Recently, a medium-level representation named "Stixel" has been extensively researched in stereo vision-based environmental perception. Obstacle detection using Stixel representation consists of three steps: static Stixel generation, dynamic Stixel generation, and Stixel segmentation. This paper focuses on the Stixel segmentation step and has two contributions. One is that it shows that Stixel segmentation performance can be enhanced by utilizing both image domain and real world domain features. The other is that it suggests that parameters used for Stixel segmentation can be effectively tuned based on genetic algorithm. The proposed method was quantitatively evaluated and the result showed that the proposed method increased Stixel segmentation accuracy compared with the previous method.

• 시스템엔지니어링학술지
• /
• 제14권2호
• /
• pp.58-66
• /
• 2018

This paper represents a systematic approach aimed at improving the performance of the proportional integral (PI) controller for the Advanced Power Reactor (APR) 1400 Feedwater Control System (FWCS). When the performance of the PI controller offers superior control and enhanced robustness, the steam generator (SG) level is properly controlled. This leads to the safe operation and increased the availability of the nuclear power plant. In this paper, a systems engineering approach is used in order to design a novel PI controller for the FWCS. In the reverse engineering stage, the existing FWCS configuration, especially the characteristics of the feedwater controller as well as the feedwater flow path to each SG from the FWCS, were reviewed and analysed. The overall block diagram of the FWCS and the SG was also developed in the reverse engineering process. In the re-engineering stage, the actual design of the feedwater PI controller was carried out using a genetic algorithm (GA). Lastly, in the validation and verification phase, the existing PI controller and the PI controller designed using GA method were simulated in Simulink/Matlab. From the simulation results, the GA-PI controller was found to exhibit greater stability than the current controller of the FWCS.

• 한국지능시스템학회논문지
• /
• 제15권2호
• /
• pp.224-229
• /
• 2005

일반적으로 유전 알고리즘은 전형적인 프로세서에서 수행할 경우 매우 큰 시간 공간 복잡도를 가진다. 따라서 유전 알고리즘 처리를 위해서는 고성능$\cdot$고가격의 프로세서를 필요로 하게 된다. 또한 이것은 유전 알고리즘을 소형 이동 로봇과 같이 비교적 간단한 룰을 필요로 하는 실제 하드웨어에 적용하는데 있어 큰 장벽으로 작용한다. 이러한 문제의 해결을 위해, 본 논문에서는 유전 알고리즘의 신속한 처리를 위해 강화된 프로세서 구조를 보인다. 정렬 네트워크와 residue number system (RNS)를 이용하여 일반적인 프로세서의 구조를 유전 알고리즘의 처리에 효율적이도록 강화할 수 있다. 정렬 네트워크는 유전 알고리즘에 필수적인 해들의 품질 비교를 하드웨어적으로 처리할 수 있게 하여 수행에 요구되는 시간을 줄일 수 있다. RNS는 산술 연산의 속도를 좌우하는 bit 사이즈를 줄여 전체적인 로직의 사이즈를 줄이고, 산술 연산의 처리 속도를 빠르게 할 수 있다.

• Journal of Mechanical Science and Technology
• /
• 제20권8호
• /
• pp.1125-1138
• /
• 2006

The optimal design of the squeeze film damper (SFD) for rotor system has been studied in previous researches. However, these researches have not been considering jumping or nonlinear phenomena of a rotor system with SFD. This paper represents an optimization technique for linear and nonlinear response of a simple rotor system with SFDs by using a hybrid GA-SA algorithm which combined enhanced genetic algorithm (GA) with simulated annealing algorithm (SA). The damper design parameters are the radius, length and radial clearance of the damper. The objective function is to minimize the transmitted load between SFD and foundation at the operating and critical speeds of the rotor system with SFD which has linear and nonlinear unbalance responses. The numerical results show that the transmitted load of the SFD is greatly reduced in linear and nonlinear responses for the rotor system.

• 전자공학회논문지B
• /
• 제33B권4호
• /
• pp.91-99
• /
• 1996

A hybrid approach employing both the genetic algorithm and the newton-raphson method is proposed for the electrical impedance computed tomography (EICT) image reconstruction. Computational experiments based on the new concept have shown promising results for several noise-free models. In particular, the resistance distribution of the tested models having resistivity ratio up to 100:1 has been reconstructed sucessfully. Using the proposed mehtod, it is also possible to get the reconstruction by the conventional iterative approaches be difficult to vonverge to a robust solution. If the compution power is enhanced further, the proposed method is expected to stimulate the practical applications of the EICT technology in the near future.

• 한국전자파학회논문지
• /
• 제11권2호
• /
• pp.177-185
• /
• 2000

The broadband EM absorbers composed of dielectrics and magnetic materials are designed. The performance of the partial initialization genetic algorithm(PIGA) is improved with three factors such as partial initialization ratios, initialization starting points and scale factors. At the frequency range over 3∼10 GHz, the optimized electromagnetic absorbers designed by using the improved PIGA. The design results obtained by enhanced PIGA's are superior to that of the using GA presented by E. Michielssen with regard to the total depth of composite materials, the depth of magnetics and maximum reflection coefficients.

• Structural Engineering and Mechanics
• /
• 제70권2호
• /
• pp.209-219
• /
• 2019

By normalizing the internal hysteresis variable and eliminating the redundant parameter, the normalized Bouc-Wen model is considered to be an improved and more reasonable form of the Bouc-Wen model. In order to facilitate application and further research of the normalized Bouc-Wen model, some key aspects of the model need to be uncovered. In this paper, hysteresis characterization of the normalized Bouc-Wen model is first studied with respect to the model parameters, which reveals the influence of each model parameter to the shape of the hysteresis loops. The parameter identification scheme is then proposed based on an improved genetic algorithm (IGA), and verified by experimental test data. It is proved that the proposed method can be an efficacious tool for identification of the model parameters by matching the reconstructed hysteresis loops with the target hysteresis loops. Meanwhile, the IGA is shown to outperform the standard GA. Finally, a simplified identification method is proposed based on parameter sensitivity, which indicates that the efficiency of the identification process can be greatly enhanced while maintaining comparable accuracy if the low-sensitivity parameters are reasonably restricted to narrower ranges.

• 한국통신학회논문지
• /
• 제27권12C호
• /
• pp.1215-1227
• /
• 2002

본 논문은 최단 경로 라우팅 문제의 해결을 위한 새로운 방식의 유전자 알고리즘(Genetic Algorithm)을 제안한다. 이를 위해 가변길이(variable-length) 염색체(chromosome) 구조와 그에 따른 유전자 부호화(genes coding) 기법을 설계하고, 부분 염색체(partial-chromosome)를 교환하는데 있어서 교차점(crossing-site)에 의존성이 없는 교배(crossover) 기법과 개체군(population)의 다양성(diversity)을 유지하는 돌연변이(mutation) 기법을 개발한다. 또한, 모든 부적합(infeasible) 염색체를 간단하게 치료할 수 있는 복구 함수(repair function)를 제안한다. 제안 교배 기법과 돌연변이 기법의 상호 동작은 제안 알고리즘이 개체군의 다양성을 유지하면서 해-표면(solution-surface)을 효과적으로 탐색할 수 있도록 하여 해의 최적성(optimality) 및 수렴(convergence) 속도의 향상을 도모한다. 제안 알고리즘에 의해 계산된 경로의 최적성은 유전자 알고리즘을 이용하는 기존의 알고리즘보다 우수하고, 수렴 속도도 빠르다는 것을 컴퓨터 시뮬레이션을 통해 확인한다. 이 결과는 대부분의 출발지와 도착지 쌍에 대해 기존의 유전자 알고리즘 기반의 최단 경로 라우팅 알고리즘에 비해 네트워크 토폴로지에 비교적 덜 민감한 것으로 나타난다.