• Journal of the Korean Data and Information Science Society
• /
• 제24권2호
• /
• pp.333-339
• /
• 2013

본 논문은 불완전계수의 모형행렬을 갖는 선형모형에서 추정가능함수를 다루고 있다. 고정효과 모형의 모수들은 일반적으로 추정가능한 모수가 아니므로 추정가능한 모수들의 함수를 구하기 위한 방법으로 완전계수의 인자분해 방법을 제시하고 있다. 완전계수의 인자분해 방법으로 구해진 추정가능함수의 타당성을 확인하기 위한 사영행렬은 불완전계수의 모형행렬을 구성하는 행벡터로 생성되는 벡터공간으로의 사영행렬과 동일함을 보여주고 있다. 완전계수의 인자분해로 추정가능함수를 구하는 방법과 모수들의 선형함수가 추정가능함수인 가의 확인을 위한 사영행렬의 이용에 관해 벡터공간의 관점에서 다루어지고 있다. 또한, 추정가능함수의 기저 구성에 관한 구체적 논의가 행해지고 있다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1993년도 하계학술대회 논문집 A
• /
• pp.306-309
• /
• 1993

Digital redesign techniques are to derive a digital controller fred the existing continuous controller and its plant so that the performance of the digitally controlled system is matched as closely as possible to that of the original continuous system at the sampling instants. Recently a new digital redesign technique, was developed to use the generalized sampled-data hold function(GSHF) based on the the grammian matrix. With this new technique the digitally redesigned system showed several good features compared with any existing technique in terms of the state match at the sampling instants, the large sampling period, and the transient reponses. This paper concerns a further improvement of the new digital redesign technique by employing the discrete grammian matrix instead of the continuous grammian matrix in defining GSHF. With this modification, it was observed by simulation that the transient response of the modified digital system shows far better performance than the digital system based on the continuous grammian matrix.

• 한국소음진동공학회논문집
• /
• 제20권10호
• /
• pp.923-930
• /
• 2010

Finite element models of dynamic systems can be updated in two stages. In the first stage, mass and stiffness matrices are updated neglecting damping, and in the second stage, damping matrices are estimated with the mass and stiffness matrices fixed. Three methods to estimate damping matrices for this purpose are proposed in this paper. The methods include one for proportional damping systems and two for non-proportional damping systems. Method 1 utilizes orthogonality of normal modes and estimates damping matrices using the modal parameters extracted from the measured responses. Method 2 estimates damping matrices from impedance matrices which are the inverse of FRF matrices. Method 3 estimates damping using the equation which relates a damping matrix to the difference between the analytical and measured FRFs. The characteristics of the three methods are investigated by applying them to simulated discrete system data and experimental cantilever beam data.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권2호
• /
• pp.297-307
• /
• 2021

Although there are several studies on the impact of pesticides use, there is no consistent conclusion about such evidence on capturing the socioeconomic independence. The propose of this paper is to investigate the economy-wide impact of pesticide use in Thailand. The research data and methodology in this paper are depended on a social accounting matrix framework incorporating the pesticide-related illness as an additional sector, following Resosudarmo and Thorbecke (1998), to explain the impact of the pesticides on the related agricultural sector, food sector, and the social welfare of different households. Thus, the main characteristics of the Thai economy can be comprehensively described by providing information contained in this framework. In this respect, the several data sets are constructed to include the economic and social structure interdependencies, which are necessary to analyze the policy implications, especially industrial policy. The results were analyzed according to the general equilibrium theory and the Leontief multiplier matrix. It reveals that the food industry and the economy are significantly affected by the pesticides. One of the most interesting findings of this paper suggest that the food sector needs to determine its output to avoid bottleneck situations and create equality across the food production system.

• 제어로봇시스템학회논문지
• /
• 제20권11호
• /
• pp.1142-1146
• /
• 2014

This paper addresses on a $H_{\infty}$ sampled-data stabilization of a Takagi-Sugeno (T-S) fuzzy system. The sampled-data stabilization problem is formulated as a discrete-time stabilization one via a direct discrete-time design approach. It is shown that the sampled-data fuzzy control system is asymptotically stable whenever its exactly discretized model is asymptotically stable. Based on an exact discrete-time model, sufficient design conditions are derived in the format of linear matrix inequalities (LMIs). An example is provided to illustrate the effectiveness of the proposed methodology.

• Journal of the Korean Data and Information Science Society
• /
• 제26권6호
• /
• pp.1557-1563
• /
• 2015

Multiclass classification is typically performed using the voting scheme method based on combining binary classifications. In this paper we propose multiclass classification method for large data, which can be regarded as the revised one-vs-all method. The multiclass classification is performed by using the hat matrix of least squares support vector machine (LS-SVM) ensemble, which is obtained by aggregating individual LS-SVM trained on each subset of whole large data. The cross validation function is defined to select the optimal values of hyperparameters which affect the performance of multiclass LS-SVM proposed. We obtain the generalized cross validation function to reduce computational burden of cross validation function. Experimental results are then presented which indicate the performance of the proposed method.

• Smart Structures and Systems
• /
• 제12권3_4호
• /
• pp.271-290
• /
• 2013

The acceleration information is significant for the structural health monitoring, which is the basic measurement to identify structural dynamic characteristics and structural vibration. The efficiency of the accelerometer is subsequently important for the structural health monitoring. In this paper, the distance measure matrix and the support level matrix are constructed firstly and the synthesized support level and the fusion method are given subsequently. Furthermore, the synthesized support level can be served as the determination for diagnosis on accelerometers, while the consensus data fusion method can be used to recover the acceleration information in frequency domain. The acceleration acquisition measurements from the accelerometers located on the real structure National Aquatics Center are used to be the basic simulation data here. By calculating two groups of accelerometers, the validation and stability of diagnosis and recovering on acceleration based on the data fusion are proofed in the paper.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2003년도 Proceeding
• /
• pp.105-110
• /
• 2003

Collaborative filtering is one of the methodologies that are most widely used for recommendation system. It is based on a data matrix of each customer's preferences of products. There could be a lot of missing values in such preference. data matrix. This incomplete data is one of the reasons to deteriorate the accuracy of recommendation system. Multiple imputation method imputes m values for each missing value. It overcomes flaws of single imputation approaches through considering the uncertainty of missing values.. The objective of this paper is to suggest multiple imputation-based collaborative filtering approach for recommendation system to improve the accuracy in prediction performance. The experimental works show that the proposed approach provides better performance than the traditional Collaborative filtering approach, especially in case that there are a lot of missing values in dataset used for recommendation system.

• 대한교통학회지
• /
• 제26권6호
• /
• pp.123-132
• /
• 2008

동적(dynamic) 기종점(origin-destination, OD) 통행량은 다양한 교통분야에 활용이 가능한데, 대표적으로 동적 통행배정모형의 입력자료와 같은 교통계획분야와 실시간 도로교통 운영분야, 그리고 교통수요 관리분야 등에도 사용할 수 있다. 이런 교통정책들을 평가하기 위해서는 정확한 동적 OD통행량의 추정은 무엇보다 중요하며, 이를 위하여 다양한 기법들이 제시되고 있다. 본 연구에서는 최근 새롭게 연구되고 있는 개인이 소지한 통행 단말기 정보를 이용하여 동적 OD통행량을 추정하고 이를 평가하고자 한다. 이를 위하여 동적 OD추정모형을 개발하고 개발된 추정모형과 동적 통행배정모형(DYNASMART-P)을 이용하여 동적 OD통행량을 추정하는데, 동적OD통행량 추정시 이용되는 단말기 정보가 표본자료(sample data)이기 때문에 이를 전수화하는 과정이 포함된다. 본 연구에서 제안한 방법으로 제주시를 대상으로 동적OD통행량을 추정한 결과, 그 가능성을 확인할 수 있었다.

• 대한교통학회지
• /
• 제21권1호
• /
• pp.53-63
• /
• 2003

관측교통량을 이용하여 추정된 OD의 오차는 관측링크집합에 크게 의존하기 때문에 얼마나 많은 수의 관측장비를 어느 지점에 위치시켜야 신뢰성을 확보할 수 있는가를 파악하는 것이 중요하다. 즉 교통량 관측지점의 위치와 개수에 따라 OD추정이 변하기 때문에 관측지점의 적정한 위치와 지점수를 산정하는 연구가 필요하다고 판단되어 본 연구는 기종점 교통수요(OD)를 추정하기 위해 교통수요추정의 정확도에 직접적으로 영향을 미치는 교통량 관측지점의 적정위치와 지점수를 선정하였다. 대안별 교통량 관측위치가 선정된 결과를 통계적인 평가지표를 이용하여 사전OD와 추정OD간의 오차를 평가하였다. 오차평가결과, 관측교통량을 이용하여 사전OD를 갱신할 때 관측교통량에 대한 관측지점의 위치 및 개수가 적정하게 반영되어 추정된 OD가 사전OD의 통행패턴을 어느 정도 잘 반영하고 있음을 알 수 있었다. 따라서 관측교통량을 이용하여 기종점간 교통수요(OD)를 추정할 때 사전OD의 통행패턴 및 특성을 잘 반영하기 위해서는 교통량 관측위치의 적정지점 및 개수선정이 중요한 요인이라고 판단된다.