• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1443-1448
• /
• 2003

Model Predictive Control (MPC) has recently found wide acceptance in industrial applications, but its potential has been much impounded by linear models due to the lack of a similarly accepted nonlinear modelling or data based technique. The authors have recently developed a new method for obtaining Volterra kernels of up to third order by use of pseudorandom M-sequence. By use of this method, nonparametric NMPC is derived in discrete-time using multi-dimensional convolution between plant data and Volterra kernel measurements. This approach is applied to an industrial polymerisation process using Volterra kernels of up to the third order. Results show that the nonparametric approach is very efficient and effective and considerably outperforms existing methods, while retaining the original data-based spirit and characteristics of linear MPC.

• 조명전기설비학회논문지
• /
• 제20권9호
• /
• pp.32-39
• /
• 2006

In this paper, an iterative subspace-based calibration algorithm for a CDMA-based antenna array in the presence of unknown gain and phase error is presented. The algorithm does not depend on the array geometry and does not require a prior knowledge of the Directions Of Arrival (DOA) of the signals. The method requires the code sequence of a reference user only. The proposed algorithm is based on the subspace method and root finding approach, and it provides estimates of the calibration vector, the DOA and the channel impulse response, by using the code sequence of a reference user. The performance of the proposed algorithm was investigated by means of computer simulations and was verified using field data measured through a custom-built W-CDMA test-bed. The data show that experimental results match well with the theoretical calibration algorithm. Also, teh study propose an efficient algorithm using the simulated annealing technique. This algorithm overcomes the requirement of initial guessing in the subspace-based approach.

• 소성∙가공
• /
• 제17권3호
• /
• pp.155-160
• /
• 2008

We analyzed a sequence of multi-stage automatic cold forging processes composed of four axisymmetric processes followed by a non-axisymmetric process using rigid-plastic finite element based forging simulators. The forging sequence selected for an example involves a piercing process and a heading process accompanying folding or overlapping, which all make it difficult to simulate the processes. To reduce computational time and to enhance the solution reliability, only the non-symmetric process was analyzed by the three-dimensional approach after the axisymmetric processes were analyzed by the two-dimensional approach. It has been emphsized that this capability is very helpful in simulating the multi-stage automatic forging processes which are next to axisymmetric or involve several axisymmetric processes.

• Mass Spectrometry Letters
• /
• 제6권3호
• /
• pp.80-84
• /
• 2015

Here, we demonstrate the use of MALDI-TOF as a fast and simple analytical approach to evaluate the DNA-binding capability of various peptides. Specifically, by varying the amino acid sequence of the peptides consisting of lysine (K) and tryptophan (W), we identified peptides with strong DNA-binding capabilities using MALDI-TOF. Mass spectrometric analysis reveals an interesting novel finding that lysine residues show sequence selective preference, which used to be considered as mediator of electrostatic interactions with DNA phosphate backbones. Moreover, tryptophan residues show higher affinity to DNA than lysine residues. Since there are numerous possible combinations to make peptide oligomers, it is valuable to introduce a simple and reliable analytical approach in order to quickly identify DNA-binding peptides.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2007년도 추계학술대회 논문집
• /
• pp.195-200
• /
• 2007

We analyzed a sequence of multi-stage automatic cold forging processes composed of four axisymmetric processes followed by a non-axisymmetric process using rigid-plastic finite element based forging simulators. The forging sequence selected for an example involves a piercing process and a heading process accompanying folding or overlapping, which all make it difficult to simulate the processes. To reduce computational time and to enhance the solution reliability, only the non-symmetric process was analyzed by the three-dimensional approach after the axisymmetric processes were analyzed by the two-dimensional approach. It has been emphsized that this capability is very helpful in simulating the multi-stage automatic forging processes which are next to axisymmetric.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2004년도 추계학술대회 및 정기총회
• /
• pp.250-278
• /
• 2004

The product rate variation problem, to be called the PRVP, is to sequence different type units that minimizes the maximum value of a deviation function between ideal and actual rates. The PRVP is an important scheduling problem that arises on mixed-model assembly lines. A surge of research has examined very interesting methods for the PRVP. We believe, however, that several issues are still open with respect to this problem. In this study, we consider convex bipartite graphs, perfect matchings, permanents and balanced sequences. The ultimate objective of this study is to show that we can provide a more efficient and in-depth procedure with a graph theoretic approach in order to solve the PRVP. To achieve this goal, we propose formal alternative proofs for some of the results stated in the previous studies, and establish several new results.

• ETRI Journal
• /
• 제27권6호
• /
• pp.713-724
• /
• 2005

This paper focuses on the problem of key-frames coding and proposes a new promising approach based on the use of fractals. The summary, made of a set of key-frames selected from a full-length video sequence, is coded by using a 3D fractal scheme. This allows the video presentation tool to expand the video sequence in a "natural" way by using the property of the fractals to reproduce the signal at several resolutions. This feature represents an important novelty of this work with respect to the alternative approaches, which mainly focus on the compression ratio without taking into account the presentation aspect of the video summary. In devising the coding scheme, we have taken care of the computational complexity inherent in fractal coding. Accordingly, the key-frames are first wavelet transformed, and the fractal coding is then applied to each subband to reduce the search range. Experimental results show the effectiveness of the proposed approach.

• Structural Engineering and Mechanics
• /
• 제34권5호
• /
• pp.597-609
• /
• 2010

A problem formulation and solution methodology for design optimization of laminated thin-walled composite beams of generic section is presented. Objective functions and constraint equations are given in the form of beam stiffness. For two different problems one for open section and the other for closed section, the objective function considered is bending stiffness about x-axis. Depending upon the case, one can consider bending, torsional and axial stiffnesses. The different search and optimization algorithm, known as Evolution Strategies (ES) has been applied to find the optimal fibre orientation of composite laminates. A multi-level optimization approach is also implemented by narrowing down the size of search space for individual design variables in each successive level of optimization process. The numerical results presented demonstrate the computational advantage of the proposed method "Evolution strategies" which become pronounced to solve optimization of thin-walled composite beams of generic section.

• 정보처리학회논문지D
• /
• 제9D권2호
• /
• pp.173-184
• /
• 2002

본 논문에서는 대용량 시퀀스 데이터베이스에 타임 워핑을 지원하는 인덱스 기반 서브시퀀스 매칭에 관하여 논의한다. 타임 워핑은 시퀀스의 길이가 서로 다른 경우에도 유사한 패턴을 갖는 시퀀스들을 찾을 수 있도록 해준다. 최근의 연구에서 타임 워핑을 지원하는 효과적인 전체 매칭 기법을 제안된바 있다. 이 기법은 데이터 시퀀스들로부터 타임 워핑에 영향을 받지 않는 특징 벡터들의 집합을 대상으로 인덱스를 구성한다. 또한, 특징 공간상에서의 필터링을 위하여 삼각형 부등식을 만족하는 타임 워핑 거리의 하한 함수를 사용한다. 본 연구에서는 이 기존의 연구에 슬라이딩 윈도우를 기반으로 하는 접두어-질의 방법을 결합하는 새로운 기법을 제안한다. 인덱싱을 위하여 각 슬라이딩 윈도우와 대응되는 서브 시퀀스로부터 특징 벡터를 추출하고, 이 특징 벡터를 인덱싱 애트리뷰트로 사용하는 다차원 인덱스를 구성한다. 질의 처리를 위하여, 조건을 만족하는 질의 접두어들에 대한 특징 벡터들을 이용하여 다수의 인덱스 검색을 수행한다. 제안된 기법은 대용량의 데이터베이스에서도 효과적인 서브시퀀스 매칭을 지원한다. 본 연구에서는 제안된 기법이 착오 기각을 유발시키지 않음을 증명한다. 제안된 기법의 우수성을 규명하기 위하여 다양한 실험을 수행한다. 실험 결과에 따르면, 제안된 기법은 실제 S&P 500 주식 데이터와 대용량의 생성 데이터 모두에 대하여 큰 성능 개선 효과를 보이는 것으로 나타났다.

• 한국수자원학회논문집
• /
• 제54권3호
• /
• pp.157-166
• /
• 2021

효율적인 댐 운영을 위해서는 높은 신뢰도를 기반으로 하는 유입량 예측이 요구된다. 본 연구에서는 최근 다양한 분야에서 사용되고 있는 데이터 기반의 예측 방법 중 하나인 딥러닝을 댐 유입량 예측에 활용하였다. 그 중 시계열 자료 예측에 높은 성능을 보이는 Sequence-to-Sequence 구조기반의 Long Short-Term Memory 딥러닝 모형(LSTM-s2s)을 이용하여 소양강 댐의 유입량을 예측하였다. 모형의 예측 성능을 평가하기 위해 상관계수, Nash-Sutcliffe 효율계수, 평균편차비율, 그리고 첨두값 오차를 이용하였다. 그 결과, LSTM-s2s 모형은 댐 유입량 예측에 대한 높은 정확도를 보였으며, 단일 유량 수문곡선 기반의 예측 성능에서도 높은 신뢰도를 보였다. 이를 통해 홍수기와 이수기에 수자원 관리를 위한 효율적인 댐 운영에 딥러닝 모형의 적용 가능성을 확인할 수 있었다.