• Communications for Statistical Applications and Methods
• /
• 제27권5호
• /
• pp.523-533
• /
• 2020

This paper suggests three available methods for finding nonnegative estimates of variance components of the random effects in mixed models. The three proposed methods based on the concepts of projections are called projection method I, II, and III. Each method derives sums of squares uniquely based on its own method of projections. All the sums of squares in quadratic forms are calculated as the squared lengths of projections of an observation vector; therefore, there is discussion on the decomposition of the observation vector into the sum of orthogonal projections for establishing a projection model. The projection model in matrix form is constructed by ascertaining the orthogonal projections defined on vector subspaces. Nonnegative estimates are then obtained by the projection model where all the coefficient matrices of the effects in the model are orthogonal to each other. Each method provides its own system of linear equations in a different way for the estimation of variance components; however, the estimates are given as the same regardless of the methods, whichever is used. Hartley's synthesis is used as a method for finding the coefficients of variance components.

• 한국산림과학회지
• /
• 제67권1호
• /
• pp.52-59
• /
• 1984

A Gentan probability q(j) is the probability that a newly planted forest will be felled at age-class j. A future change in growing stock and yield of the forests can be predicted by means of this probability. On the other hand a state of the forests is described in terms of an n-vector whose components are the areas of each age-class. This vector, called age-class vector, flows in a n-1 dimensional simplex by means of $n{\times}n$ matrices, whose components are the age-class transition probabilities derived from the Gentan probabilities. In the simplex there exists a fixed point, into which an arbitrary forest age vector sinks. Theoretically this point means a normal state of the forest. To each age-class-transition matrix there corresponds a single normal state; this means that there are infinitely many normal states of the forests.

• 한국정보처리학회논문지
• /
• 제3권1호
• /
• pp.167-180
• /
• 1996

스칼라 처리지향의 기계에서 벡터 처리를 위해서는 스칼라 처리가 벡터 요소 수 만큼 수행되어야 한다. 소위 von Neumann원리에 의한 벡터 처리기법이다. 메모리를 악세스 하는 장치로는 명령어의 순차적 계수를 위한 프로그램 계수기 뿐이기 때문에 벡터 데이터의 악세스는 명령어의 지시나 또는 ALU 의 주소 계산에 의해 수행되어 야 한다. 여기서는 재래식 개념의 하드웨어적 결합을 보충하기 위해 벡터 요소들을 입체적으로 악세스하기 위한 악세스 장치의 설계를 제안한다. 벡터의 구조 처리를 위한 필요성은 명령어군에 포함되었고 그들 명령어들은 데이터 처리와 동시에 데이터 악세스 안에 처리되도록 한다.

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 2000년도 전력전자학술대회 논문집
• /
• pp.181-185
• /
• 2000

To get high efficiency in variable speed control of induction motor it is required that the vector control should be separated from flux components current and torque component current. In this paper the vector control is modeled by the estimation of the stator flex. Representing induction motor speed controller as a digital system with he use of he 32bit DSP improves the motor control performance The IGBT is used as the switching device and the validity of the proposed vector control is proved through voltage current wave and the characteristics of the velocity response as the drive circuit being simplified

• 말소리와 음성과학
• /
• 제6권4호
• /
• pp.127-132
• /
• 2014

This paper describes a robust speech recognition technique by reconstructing spectral components mismatched with a training environment. Although the cluster-based reconstruction method can compensate the unreliable components from reliable components in the same spectral vector by assuming an independent, identically distributed Gaussian-mixture process of training spectral vectors, the presented method exploits the temporal dependency of speech to reconstruct the components by introducing a hidden-Markov-model prior which incorporates an internal state transition plausible for an observed spectral vector sequence. The experimental results indicate that the described method can provide temporally consistent reconstruction and further improve recognition performance on average compared to the conventional method.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 추계학술대회 논문집 전력기술부문
• /
• pp.60-62
• /
• 2001

This paper shows the optimal bidding strategy determination method using Nash equilibrium concept by defining bidding strategy vector. This vector is 2-dimension vector whose components are generation amount and generation cost. Thereby we are able to make all possible strategies and their's payoff table. And then we erase dominated strategies one by one so that we obtain Nash equilibrium, the optimal bidding strategy of generation bidding game.

• 한국멀티미디어학회논문지
• /
• 제14권2호
• /
• pp.210-218
• /
• 2011

일반적으로 GIS 디지털 법의 표현 및 저장 방식에는 아스키(ASCII)와 바이너리(binary) 형식이 있다. 이들 중 대부분 GIS 응용 분야에서는 대용량의 맵 데이터 전송을 위하여 바이너리 형식의 벡터 맵 데이터를 주로 사용한다. 본 논문에서는 다양한 정밀도를 갖는 바이너리 형식의 벡터 맵 데이터의 효율적인 전송 및 저장을 위하여 벡터 맵 데이터의 주요 레이어를 표현하는 폴리라인 및 폴리곤 성분에 대한 계층적 압축 기법을 제안한다. 제안한 기법에서는 정밀 벡터 맵 데이터의 무손실 압축을 위하여 공간 영역 상에서 에너지 집중(energy compaction)을 수행하며, 64bit 부동소수점 좌표에 대하여 정수부와 소수부를 독립적으로 압축 부호화를 수행한다. 실험 결과로부터 제안한 압축 기법이 최소 200% 이상, 최대 900% 이상의 압축 효율을 나타냄을 확인하였으며, 기존의 데이터 압축 기법인 7z, zip, rar 및 gz에 비히여 우수한 압축률을 가지는 것을 확인하였다.

• 지구물리와물리탐사
• /
• 제14권4호
• /
• pp.298-304
• /
• 2011

한국해양연구원에서는 2011년 4월에 쇄빙선 아라온호, Oceaneering사의 ROV (Remotely Operated Vehicle) 및 삼성분자력계(Three component vector magnetometer)를 이용하여 남서태평양 라우분지에 위치한 TA25해산 칼데라 서측 사면에 대하여 심해자력탐사를 실시하였다. 연구지역은 약 900 m ~ 1200 m의 수심대역을 보인다. 이 심해탐사를 위하여 국내 최초로 3000 m급 심해 삼성분자력계를 제작하였다. 자력계센서부분은 ROV의 상부에, 자력계로거부분은 ROV의 하부에 장착하였으며, ROV는 ROV의 정확한 위치를 알려주는 USBL (Ultra Short Base Line)과 고도계를 이용하여 정해진 측선을 따라 해저면에서 약 25 ~ 30 m의 수직 간격을 유지하며 이동하였다. 삼성분자력계는 자기장의 X(북쪽성분), Y(동쪽성분)과 Z(수직성분)의 벡터성분을 측정하므로 모션센서를 이용하여 ROV의 움직임(피치, 롤, 요)에 대한 보정을 실시하였다. 자력센서자료와 모션센서자료는 ROV의 광케이블과 아라온호의 네트워크를 이용하여 노트북에 저장되며 ROV의 정확한 위치자료인 USBL자료는 후처리를 통하여 병합하였다. 이렇게 획득한 심해 삼성분자력자료는 조사지역의 유망한 열수광상을 유추하는데 유용하게 활용될 것으로 판단된다.

• Journal of Power Electronics
• /
• 제12권3호
• /
• pp.477-486
• /
• 2012

This paper proposes a novel current control scheme for vector-controlled induction machine (IM) drives in the overmodulation (OVM) range, with which the voltage utilization of the voltage-source inverter (VSI) can be maximized. In the OVM region, the original voltage reference is modified by changing its magnitude and angle, which causes the motor current to be distorted, resulting in a deterioration of the current control performance. To meet with this situation, the harmonic components in the feedback currents should be eliminated before being input to the PI current controllers. For this, a composite observer is applied to extract the fundamental and harmonic components from the distorted currents, which gives a good performance without a delay and the effect of a fundamental frequency variation. In addition, through a detailed analysis of the response of the PI current controllers in the OVM range, the effectiveness of using the composite observer is demonstrated. Simulation and experimental results for a 3-kW induction motor drive are shown to verify the validity of the proposed method.

• Journal of the Korean Data and Information Science Society
• /
• 제27권2호
• /
• pp.549-557
• /
• 2016

In this paper we propose multiclassification method for large data sets by ensembling least squares support vector machines (LS-SVM) with principal components instead of raw input vector. We use the revised one-vs-all method for multiclassification, which is one of voting scheme based on combining several binary classifications. The revised one-vs-all method is performed by using the hat matrix of LS-SVM ensemble, which is obtained by ensembling LS-SVMs trained using each random sample from the whole large training data. The leave-one-out cross validation (CV) function is used for the optimal values of hyper-parameters which affect the performance of multiclass LS-SVM ensemble. We present the generalized cross validation function to reduce computational burden of leave-one-out CV functions. Experimental results from real data sets are then obtained to illustrate the performance of the proposed multiclass LS-SVM ensemble.