• Journal of the Korean Data and Information Science Society
• /
• 제9권1호
• /
• pp.63-70
• /
• 1998

The singular value decomposition is a tool which is used to find a linear structure of reduced dimension and to give interpretation of the lower dimensional structure about multivariate data. In this paper the singular value decomposition is reviewed from both algebraic and geometric point of view and, is illustrated the way which the tool is used in the multivariate techniques finding a simpler geometric structure for the data.

• 전력전자학회논문지
• /
• 제9권4호
• /
• pp.364-371
• /
• 2004

본 논문에서는 PWM 펄스를 랜덤하게 위치시킬 수 있는 새로운 3상 공간 벡터 RPWM(Random PWM) 기법을 제안한다. 기존의 RPWM 방식과 달리 3상 대칭 변조 기법에 의해 발생된 공간 전압 벡터 PWM(Space Vector PWM) 펄스를 랜덤하게 위치시키는 원리를 각 섹터마다 달리 적용하여 RPWM을 구현하는 기법을 제안한다. 제안 된 기법의 타당성 검증을 위하여, 제안된 RPWM 기법 알고리즘에 대해 MATLAB/SIMULINK를 이용한 시뮬레이션 및 DSP를 이용한 IGBT 인버터를 구성하여 실험을 수행한다. 특히 10,000 rpm 이상의 고속운전 시, 공간 전압 벡터 PWM 기법과 제안된 RPWM 기법에 의해 발생되는 인버터 출력 전류의 고조파 스펙트럼을 비교한다. 즉 전류 고조파 스펙트럼에서 스위칭 주파수 정수배에 해당하는 고조파의 크기를 비교하여 제안된 RPWM의 기법의 타당성을 검증하고자한다.

• 대한안전경영과학회지
• /
• 제10권3호
• /
• pp.43-53
• /
• 2008

This study is carried out for the fire safety of the factory building, the fire risk reduction measure in compliance with an example approached in fire risk reduction systematically, contribute to reduce the fire risk. The analytical fire risk process of discovering, identifying, estimating and evaluating risk and control measure as risk reduction measures are core concept, applies loss prevention with loss control techniques. The painting process in the workplace where the fire hazard and death accident accompanies coexists. Loss prevention problem of creation prevention of dangerous atmosphere at workplace is health and human services problem of normal circumstances, must be inspected with problem of combustible gases at the time of fire explosion. Static electricity measure accomplished the risk control process thoroughly as the fire risk reduction process model with the ignition sources measure which is presented. Fire risk from within organizing will be able to classify with each field by detailedly but risk treatment process will be able to apply basically all the same concept. Consequently about risk management example from before, this study is proposed risk management techniques that standardized rightly in the actual condition of organization with one plan, with discovery of fire risk, the feedback process in compliance with a fire risk reduction and the review which control the result is joint responsibility of engineer, technical expert and manager as part of safety management to practice with the fact must be supervised.

• International Journal of Control, Automation, and Systems
• /
• 제6권2호
• /
• pp.180-185
• /
• 2008

A new method for relative error continuous and discrete time model order reduction is proposed. The reduction technique is based on two recently developed methods, namely frequency domain balanced truncation within a frequency bound and inner-outer factorization techniques. The proposed method is of interest for practical model order reduction because in this context it shows to keep the accuracy of the approximation as high as possible without sacrificing the computational efficiency. Numerical results show the accuracy and efficiency enhancement of the method.

• 복합신소재구조학회 논문집
• /
• 제7권1호
• /
• pp.45-52
• /
• 2016

There are only 10 projects of the domestic greenhouse gas(GHG) emissions trading scheme in building sector (i.e., 1.5% of 652 registered projects) because the certified methodologies to reduce GHG emissions can not be applied to building sector. This study presents remodeling techniques to reduce GHG emissions in existing buildings. First of all, preconditions and related regulations were reviewed. And then, a pool of factors for GHG reduction are selected and evaluated with respect to factors for reducing energy consumption. This study also investigates the criteria and the decision making process for remodeling techniques to reduce GHG emissions. Finally, the remodeling techniques using the decision making process were grouped based on redundancy of each effect. If reducing methodologies for GHG offset program can be developed using the analyzed remodeling techniques in this study, registered projects in building sector would be increase.

• 전기학회논문지
• /
• 제57권11호
• /
• pp.1962-1969
• /
• 2008

According to the increase of power demand and expansion of downtown, it is necessary to install transformers additionally in operating substations and construct substations in residential area. But the public complaint is increased due to the transformer noise of the substation. KEPCO has used a vibration preventing pad, various soundproof walls and an encloser to transformers in outdoor substations, and a soundproof door, shutter and wind-path soundproof equipment in indoor substations to block the sound propagation of the transformers. But these noise reduction methods are not satisfied. It should be considered to reduce transformer noise itself. In this paper, we investigated core noise reduction techniques to develope a low noise transformer. The techniques to reduce core noise of the transformer are application of high permeability grain oriented silicon sheets, decrease of magnetic flux density of core, application of 6step-lap core stacking method, improvement of core binding method(binding addition, band fixing) and application of rubber damper in oder to reduce transmission of core vibration, etc.

• Journal of Multimedia Information System
• /
• 제6권4호
• /
• pp.225-234
• /
• 2019

Deep learning has become increasingly popular in both academic and industrial areas nowadays. Various domains including pattern recognition, Computer vision have witnessed the great power of deep neural networks. However, current studies on deep learning mainly focus on quality data sets with balanced class labels, while training on bad and imbalanced data set have been providing great challenges for classification tasks. We propose in this paper a method of data analysis-based data reduction techniques for selecting good and diversity data samples from a large dataset for a deep learning model. Furthermore, data sampling techniques could be applied to decrease the large size of raw data by retrieving its useful knowledge as representatives. Therefore, instead of dealing with large size of raw data, we can use some data reduction techniques to sample data without losing important information. We group PCB characters in classes and train deep learning on the ResNet56 v2 and SENet model in order to improve the classification performance of optical character recognition (OCR) character classifier.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2011년도 춘계학술대회 논문집
• /
• pp.1750-1756
• /
• 2011

After Kyoto Protocol was adopted for green gas reduction, each nations are stepping up efforts to reduce $CO_2$ of a typical green gas. Construction industry also is trying $CO_2$ reduction with the techniques of two types which are software and hardware techniques. The software technique are Passive Design considered green gas emission and the environment impact assessment by LCA. The hardware techniques are adjustment of equipment system and development of eco- friendly material. But, it is nonexistent that a study related to $CO_2$ emission considered detail process in construction industry. This study analyzed the relativeness of equipment combination and $CO_2$ emission by calculate $CO_2$ emission follow to equipment combination on earth-work which is the process emitted most $CO_2$ among railway bedding construction.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.595-598
• /
• 2002

This paper proposes neighborhood reduction techniques in local search of the customer decomposition subproblem in the Multi Depots Vehicle Routing Problem with Time Windows (MDVRPTW) by using geospatial relation among depots and customers. The neighborhood of the customer decomposition subproblem can be simply and well defined but it should include excessively bad solution candidates. Our techniques find such candidates by using information of the problem domain, geographical relation. We use our techniques in Tabu Search and evaluate the effectiveness in computer experiment.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 추계학술대회
• /
• pp.83-86
• /
• 1997

To provide ameliorated aided environment to hearing impaired listeners, background noise reduction techniques are investigated as a front-end of conventional hearing aids, and their effects are tested in a subjective manner. Several speech enhancement schemes were implemented and preference tests or normal listeners are performed to select the best possible scheme or hearing impaired listeners. Results indicated that SDT scores without the speech enhancement scheme drop more sharply as SNR decreases than those with the speech enhancement techniques. SDT scores obtained or hearing impaired listeners with hearing aids showed large variability. However, all impaired listeners preferred noise suppressed sounds to unsuppressed ones.