• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.5
• /
• pp.593-597
• /
• 2004

This paper presents the new design method of fuzzy control system for nonlinear system. Many conventional design methods for fuzzy controller find the control gain for stabilizing fuzzy controller with some mathematical approaches. However, there exist some controllers which are hard to design with mathematical approach. In order to solve these problems, we propose the intelligent design method for fuzzy controller by using genetic algorithm with evolution strategy. The genetic algorithm with evolution strategy finds the control gain by changing the evolution region of chromosome. Finally, an application example of stabilizing a cart-pole typed inverted pendulum system will be given to show the stabilizability of the fuzzy controller.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.6
• /
• pp.99-108
• /
• 1998

In this paper, we proposed a multi-layer neural network for direct control method of nonlinear system. The proposed control method uses neural network as the controller to learn inverse model of plant. The neural network used consists of two parts; one part is for identification of linear part, and the other is for identification of nonlinear part of inverse system. The neural network has to be learned the liner part with RLS algorithm and the nonlinear part with error of plant. From the simulation and experiment of tracking control to use one link manipulator as plant, we proved usefulness of the proposed control method to comparing to conventional direct neural network control method. By comparing the two methods, from simulation and experiment, we were convinced that the proposed control method is more simple and accuracy than the conventional method. Moreover, number of weight and bias to be controller parameter are small, and it has smaller steady state error than conventional method.

• Korean Journal of Environment and Ecology
• /
• v.28 no.5
• /
• pp.523-530
• /
• 2014

One juvenile raptor which was not able to be identified due to its head damage was discovered on a roadside in Janggok-ri, Jori-eup, Paju on 28th June, 2011. The species was identified by DNA barcoding. After polymerase chain reaction (PCR) of the mitochondrial cytochrome c oxidase subunit I gene (COI), we obtained 695 bp sequences. We analyzed the obtained COI sequence with similar sequences from the BOLD systems and BLAST of the NCBI Genbank, and discovered that its sequence showed 100 % similarity values with the one of the five gray-faced buzzards which were previously researched. In addition, it was confirmed to be a female through sex determination using DNA. Such results are important information as it confirms the breeding of the gray-faced buzzards for the first time in 43 years since its breeding was last recorded in 1968, in Paju. Wildlife rescue center needs to work with adjacent consigned registration and preservation institutions when carcass of wild animals is collected or DNA samples are obtained for more accurate both species and sex identification through a systematic management system in the future. Furthermore, the obtained DNA sample of the gray-faced buzzard and COI gene, DNA barcode, could be used as reference standards for similar researches in the future.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.531-536
• /
• 2005

The purpose of this paper is to improve the hysteresis characteristics of a stack type piezoelectric actuator using system identification and tracking control. Recently, several printing methods that cost less and are faster than previous semiconductor processes have been developed for the production of electric paper and RFID. The system proposed in this study prints by spraying the molten metal, and consists of a nozzle, heating furnace, operating actuator, and an XYZ 3-axis stage, As an operating system, the piezoelectric(PZT) method has very valuable uses. However, the PZT actuator has a very big hysteresis characteristic due to the ferroelectric characteristics of the PZT element. This causes problems in the system position control characteristics and deteriorates the performance of the system. In this study, an investigation was conducted to improve the hysteresis characteristics of the PZT actuator that has an output displacement for the input voltage. The study proposed a inverse hysteresis model, a mathematic modeling method that can express the geometric relationship between voltage and displacement, in order to reduce the hysteresis of the PZT actuator. In addition, system identification and PID control methods were examined. Also, it was confirmed that the proposed control strategy gives good precision position control performance.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.6
• /
• pp.789-799
• /
• 1999

In this paper, the optimization of fuzzy inference systems is proposed for fuzzy model of nonlinear systems. A fuzzy model needs to be identified and optimized by means of the definite and systematic methods, because a fuzzy model is primarily acquired by expert's experience. The proposed rule-based fuzzy model implements system structure and parameter identification using the HCM(Hard C-mean) clustering method, genetic algorithms and fuzzy inference method. Two types of inference methods of a fuzzy model are the simplified inference and linear inference. in this paper, nonlinear systems are expressed using the identification of structure such as input variables and the division of fuzzy input subspaces, and the identification of parameters of a fuzzy model. To identify premise parameters of fuzzy model, the genetic algorithms is used and the standard least square method with the gaussian elimination method is utilized for the identification of optimum consequence parameters of fuzzy model. Also, the performance index with weighting factor is proposed to achieve a balance between the performance results of fuzzy model produced for the training and testing data set, and it leads to enhance approximation and predictive performance of fuzzy system. Time series data for gas furnace and sewage treatment process are used to evaluate the performance of the proposed model.

• Journal of the Korean Wood Science and Technology
• /
• v.24 no.1
• /
• pp.87-94
• /
• 1996

한국산 동백나무겨우살이(Pseudixus japonicus) 추출물의 암치료를 위한 생약으로서 활성 유효성을 검증하기 위하여 메탄올, 석유에테르, 클로로포름, 초산 에틸 용매로 순차적으로 추출하여 겨우살이 추출물의 다섯 가지 분획을 얻어, 이에 대하여 in vitro로 1차와 2차 검색 시스템을 사용해 항암활성 성분을 체계적으로 검색하였다. 다섯 가지 분획 중 클로로포름 가용성 분획이 1차 검색 세포인 $P388D_1$에 대해 가장 높은 항암활성을 나타내어 MSB1, NIH/3T3, SNU-1, SNU-C2A 등 2차 검색 시스템에 대해 클로로포름 가용성 분획의 항암활성을 다양한 농도하에서 비교 검색하였다. 혈액암 세포중 특히 $P388D_1$의 생장이 클로로포름 추출물에 의해 강하게 저해되었으며, 형질전환된 생쥐의 태아 섬유아세포와 사람의 대장암, 위암세포들도 어느 정도의 생육저해를 나타내었다. 이 클로로포름 가용성 분획의 주성분은 원소분석, 발색시약과의 반응, IR, GC-MS, $^{13}C$-NMR의 스펙트럼의 결과로 세 종류의 알칼로이드 화합물로 확인되었고, 부성분으로는 지방산 메틸 에스테르와 프탈라이드 화합물이 MS 스펙트럼을 통해 동정되었다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.3
• /
• pp.188-193
• /
• 2000

This paper presents an approach to obtain a reduced neuro-fuzzy model for a plant. The Neuro-Fuzzy Network are compose of the Radial Basis Function Networks with Gausis membership and learned by using temporal back propagation. The dependency in rough set theory is used to eliminate rules. Dependency between the condition membership value of each rule in a model and the output of the plant can allow us to see how much contribution the rule is to identify the plant. While the reduced model maintains the same performance as the original one, the selection algorithm can minimize its complexity and redundancy of the structure.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.737-742
• /
• 2003

In this paper, we propose a new wavelet-based fuzzy modeling using a DNA coding method. Generally, it is well known that the DNA coding method is more diverse in the knowledge expression and better in the optimization performance than the genetic algorithm (GA) because it can encode more plentiful genetic information based on the biological DNA. The proposed method makes a fuzzy model by using the wavelet transform, in which coefficients are identified by the DNA coding method. Thus we can effectively get the fuzzy model of nonlinear system by using the advantages of both wavelet transform and DNA coding method. In order to demonstrate the superiority of the proposed method, it is compared with the GA.

• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.680-682
• /
• 2018

기후변화에 따라 매개 질병의 발병 빈도가 증가하고 있으며 모기와 같은 매개체에 의해 전염되는 매개 질병은 인구집단에 대한 중요한 위협 요인이다. 이러한 질병 관리를 위해 지역별 모기 서식 현황을 모니터링 하는 시스템의 필요성이 강조되고 있다. 하지만 현재의 모기 모니터링은 개체 파악을 위한 분류와 동정을 사람이 직접 수행하기에 오랜 시간이 소요된다. 이 연구는 그러한 문제점을 해결하고 미래 매개곤충 서식 현황 파악 시스템의 기반을 마련하기 위해 심층 신경망(Deep Neural Networks)을 활용하여 한국 주요 매개모기 종 분류를 수행하고 결과를 분석하였다. 종 분류를 위한 모델은 잘 알려진 신경망 모델인 DenseNet(Densely Connected Networks)을 사용하였고 이를 직접 촬영한 모기 데이터와 약간의 변형을 가한 모기 데이터를 사용하여 학습시켰다. 학습 데이터를 각각 5배, 20배, 100배로 증강하여 실제 데이터의 부족을 보완하였으며, 이를 통해 최대 99.48%의 정확도를 달성하였다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.11a
• /
• pp.284-293
• /
• 2007

본 논문에서는 실시간 동정 환경에 효과적인 L-CAA 에이전트 구조를 제안한다. L-CAA 에이전트 구조는 변화하는 환경에 대한 적응성을 높이기 위해, 선행 연구를 통해 개발된 행위 기반 에이전트 구조인 CAA에 강화 학습 기능을 추가하여 확장한 것이다. 안정적인 성능을 위해 L-CAA에서 행위 선택 메커니즘은 크게 두 단계로 나뉜다. 첫 번째 단계에서는 사용자가 미리 정의한 각 행위의 수행 가능 조건과 효용성을 검사함으로써 행위 라이브러리로부터 실행할 행위들을 추출한다. 하지만 첫 번째 단계에서 다수의 행위가 추출되면, 두 번째 단계에서는 강화 학습의 도움을 받아 이들 중에서 실행할 하나의 행위를 선택한다. 즉, 강화 학습을 통해 갱신된 각 행위들의 Q 함수 값을 서로 비교함으로써, 가장 큰 기대 보상 값을 가진 행위를 선택하여 실행한다. 또한 L-CAA에서는 실행 중인 행위의 유지 가능 조건을 지속적으로 검사하여 환경의 동적 변화로 인해 일부 조건이 만족되지 않는 경우가 발생하면 현재 행위의 실행을 즉시 종료할 수 있다. 그 뿐 아니라, L-CAA는 행위 실행 중에도 효용성이 더 높은 다른 행위가 발생하면 현재의 행위를 일시 정지하였다가 복귀하는 기능도 제공한다. 본 논문에서는 L-CAA 구조의 효과를 분석하기 위해, 대표적인 동적 가상환경인 Unreal Tournament 게임에서 자율적을 동작하는 L-CAA기반의 UTBot 들을 구현하고, 이들을 이용하여 성능실험을 전개해본다.