• Nuclear Engineering and Technology
• /
• 제33권5호
• /
• pp.483-497
• /
• 2001

In this work, a neuro-fuzzy inference system combined with the wavelet denoising, PCA (principal component analysis) and SPRT (sequential probability ratio test) methods is developed to detect the relevant sensor failure using other sensor signals. The wavelet denoising technique is applied to remove noise components in input signals into the neuro-fuzzy system The PCA is used to reduce the dimension of an input space without losing a significant amount of information. The PCA makes easy the selection of the input signals into the neuro-fuzzy system. Also, a lower dimensional input space usually reduces the time necessary to train a neuro-fuzzy system. The parameters of the neuro-fuzzy inference system which estimates the relevant sensor signal are optimized by a genetic algorithm and a least-squares algorithm. The residuals between the estimated signals and the measured signals are used to detect whether the sensors are failed or not. The SPRT is used in this failure detection algorithm. The proposed sensor-monitoring algorithm was verified through applications to the pressurizer water level and the hot-leg flowrate sensors in pressurized water reactors.

• 한국정보처리학회논문지
• /
• 제7권4호
• /
• pp.1044-1049
• /
• 2000

In detection of a scene change of the moving pictures which has massive information capacity, the temporal sampling method has faster searching speed than the sequential searching method for the whole moving pictures, yet employed searching algorithm and detection interval greatly affect searching time and searching precision. In this study, the whole moving pictures were primarily retrieved by the temporal sampling method. When there exist a scene change within the sampling interval, we suggested a fast searching algorithm using binary searching and derived an equation formula to determine optimal primary retrieval which can minimize computation, and showed the result of the experiment on MPEG moving pictures. The result of the experiment shows that the searching speed of the suggested algorithm is maximum 13 times faster than the one of he sequential searching method.

• 한국농공학회지
• /
• 제37권1호
• /
• pp.90-99
• /
• 1995

In this study, the algorithm for the optimum design of cantilever retaining wall was de veloped and solved using Modified Method of Feasible Directions(MMFD), Sequential Linear Programming(SLP) and Sequential Quadratic Programming(SQP). The algorithm was applied to the optimum design of 3-different height cantilever re tairing walls. It was shown that even though the starting points and optimization strategies are dif- ferent, the objective function and optimum design variables converge to within a close range, and consequently the reliability and efficiency of the underlying optimum design algorithm can be verified. It is expected that the optimum design algorithm developed in this study can be utilized efficiently for the optimum design of any scale cantilever retaining wall. Using optimum design method, cantilever retaining wall will be designed more economi- cally and reasonably than using traditional design method.

• 대한원격탐사학회지
• /
• 제26권6호
• /
• pp.681-691
• /
• 2010

LiDAR waveform decomposition plays an important role in LiDAR data processing since the resulting decomposed components are assumed to represent reflection surfaces within waveform footprints and the decomposition results ultimately affect the interpretation of LiDAR waveform data. Decomposing the waveform into a mixture of Gaussians involves two related problems; 1) determining the number of Gaussian components in the waveform, and 2) estimating the parameters of each Gaussian component of the mixture. Previous studies estimated the number of components in the mixture before the parameter optimization step, and it tended to suggest a larger number of components than is required due to the inherent noise embedded in the waveform data. In order to tackle these issues, a new LiDAR waveform decomposition algorithm based on the sequential approach has been proposed in this study and applied to the ICESat waveform data. Experimental results indicated that the proposed algorithm utilized a smaller number of components to decompose waveforms, while resulting IMP value is higher than the GLA14 products.

• Journal of Electrical Engineering and Technology
• /
• 제12권4호
• /
• pp.1675-1682
• /
• 2017

This paper deals with three dimensional orientation estimation algorithm for an attitude and heading reference system (AHRS) based on nine-axis inertial/magnetic sensor signals. In terms of the orientation estimation based on the use of a Kalman filter (KF), the quaternion is arguably the most popular orientation representation. However, one critical drawback in the quaternion representation is that undesirable magnetic disturbances affect not only yaw estimation but also roll and pitch estimations. In this paper, a sequential direction cosine matrix-based orientation KF for AHRS has been presented. The proposed algorithm uses two linear KFs, consisting of an attitude KF followed by a heading KF. In the latter, the direction of the local magnetic field vector is projected onto the heading axis of the inertial frame by considering the dip angle, which can be determined after the attitude KF. Owing to the sequential KF structure, the effects of even extreme magnetic disturbances are limited to the roll and pitch estimations, without any additional decoupling process. This overcomes an inherent issue in quaternion-based estimation algorithms. Validation test results show that the proposed method outperforms other comparison methods in terms of the yaw estimation accuracy during perturbations and in terms of the recovery speed.

• 한국농공학회지
• /
• 제36권2호
• /
• pp.111-122
• /
• 1994

In this study, an algorithm which can be applied to the optimum design of simple steel plate girders was developed, and efficient optimization strategies for the solution of algorithm were found out. The optimum design algorithm consists of 3-levels of optimization. In the first and second levels of optimization, the absolute maximum bending moment and shearing force are extracted and in the third level of optimization, the optimum cross section of steel plate girder is determined. For the optimum design of cross section, the objective function is formulated as the total area of cross section and constraints are derived in consideration of the various stresses and the minimum dimension of flange and web based on the part of steel bridge in the Korea standard code of road bridge. Sequential unconstrained minimization technique using the exterior penalty function method(SUMT-EP), sequential linear programming(SLP) and sequential quadratic programming (SQP) are proved to be efficient and robust strategies for the optimum design of simple plate girder cross section. From the reliable point of view, SLP is the most efficient and robust strategy and SQP is the most efficient one from the viewpoint of converguency and computing time.

• 한국통신학회논문지
• /
• 제36권7B호
• /
• pp.818-826
• /
• 2011

본 논문은 플래쉬 메모리 접근 시 소거 횟수를 줄이기 위해서 섹터 매핑 기법을 바탕으로 지역성(Locality)을 FTL (Flash Translation Layer) 알고리듬을 제안한다. 섹터 매핑 기법은 다른 기존의 매핑 기법보다 성능은 우수하나, 매핑 테이블이 커지는 단점을 가진다. 제안한 알고리듬에서는 동적 테이블 업데이트 방식으로 매핑 테이블의 크기를 줄였으며, 순차적 지역성 (Sequential Locality)을 이용하여 순차쓰기를 처리하고 임의쓰기에서 hot 섹터를 추출하였다. 실험결과 제안된 알고리듬은 기존의 BAST, FAST, 섹터 매핑 알고리듬과 비교하여 75.2%, 65.8%, 10.3%의 소거 횟수가 감소함을 보인다.

• Korea-Australia Rheology Journal
• /
• 제12권1호
• /
• pp.83-92
• /
• 2000

In resin transfer molding (RTM) process, preplaced fiber mat is set up in a mold and thermoset resin is injected into the mold. An important interest in RTM process is to minimize cycle time without sacrificing part quality or increasing cost. In this study, the numerical simulation and optimization process in filling stage were conducted in order to determine the optimum gate locations. Control volume finite element method (CVFEM) was used in this numerical analysis with the coordinate transformation method to analyze the complex 3-dimensional structure. Experiments were performed to monitor the flow front to validate simulation results. The results of numerical simulation predicted well the experimental results with every single, simultaneous and sequential injection procedure. We performed the optimization analysis for the sequential injection procedure to minimize fill time. The complex geometry of an automobile bumper core was chosen. Genetic algorithm was used in order to determine the optimum gate locations with regard to 3-step sequential injection case. These results could provide the information of the optimum gate locations in each injection step and could predict fill time and flow front.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권2호
• /
• pp.75-84
• /
• 2005

연속패턴은 다양한 분야에서 사용되는 데이타 마이닝 기법의 한 종류이다. 하지만 현재의 연속 패턴 방법은 한개의 속성내에서의 패턴만을 감지할 수 있으며 속성간의 패턴을 생성할 수 없다. 다차원의 연속패턴은 일차원에 비하여 훤씬 유용한 정보를 제공할 수 있다. 본 연구에서는 Hellinger 엔트로피 함수를 사용하여 다차원의 연속패턴을 생성하는 방법을 게시한다 기존의 연속패턴방법과 달리 본 방법에서는 각 연속패턴의 중요도를 자동으로 계산할 수 있다. 또한 계산의 복잡도를 감소시키기 위한 다수의 법칙이 개발되었으며 다수의 실험 결과를 제시하였다.

• 한국해양공학회:학술대회논문집
• /
• 한국해양공학회 2004년도 학술대회지
• /
• pp.67-72
• /
• 2004

Many port states such as New Zealand, the USA, Australia and Canada have strict regulations to prevent ships which arrive in their port from discharging polluted ballast water which contain harmful aquatic organism and pathogens. They are notified that transfer of polluted ballast water can cause serious injury to public health and damage to property and environment. For this reason, they perceived that the ballast exchange in deep sea is the most effective method, together with submitting the ballast management plan which contains the effective exchange method, ballast system and safety consideration. In this study, we pursued both nautical engineering analysis and optimization of algorithm in order to generate the sequence of stability and rapidity. Heuristic Algorithm was chosen on the basis of optimality and applicability to a sequential exchange problem. We have built an optimized algorithm, for automatic exchange of ballast water, by redefining core elements of the $A^\ast$ algorithm, such as node, operator and evaluation function. Final version of the optimized algorithm has been applied to existing bulk carrier and the performance of the algorithm has been verified successfully.