• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.591-592
• /
• 2008

본 논문은 배터리를 이용하는 시스템의 사용시간을 극대화하기 위하여 두 가지 해결책을 제시한다. 첫 번째, 우리는 멀티 프로세서 시스템에서 Dynamic Voltage Scaling(DVS)을 이용하여 에너지 소모를 최소화시킨다. 다른 어프로치와의 큰 차이점은 테스크의 실행 시간을 deadline까지 확장시켜 에너지 소모를 최소화할 뿐만 아니라 테스크의 실행 사이클 수가 감소할것을 고려하여 테스크를 나누어 다른 동작 주파수를 적용 시키고 이를 수학적 방법으로 도출한다. 두 번째, 배터리의 discharge 특성인 capacity rate effect와 recovery effect를 고려하여 프로세서들의 에너지 소모 프로파일을 재구성함으로서 배터리 라이프타임을 최적화시킨다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.54-55
• /
• 2018

본 논문은 지속적으로 수집되는 전력량계 데이터를 자동으로 처리, 분석하기 위한 IoT 데이터 기반 자동분석 기법을 제시한다. 에너지 효율을 높이기 위해서는 대상 설비의 관리, 모니터링을 통해 운영을 최적화해야 한다. IoT 기술을 이용하여 에너지 설비 사용 효율을 확인하고, 관리 여부를 판단하는 진단기술을 구현하기 위해서는, IoT 전력량계를 통해 수집된 데이터를 다양한 머신러닝 알고리즘에 입력하여 관리에 필요한 결과 지표를 도출할 수 있어야 한다. 이런 기능을 제공하는 IoT 수집 시스템의 모니터링 및 자동 진단 시스템은 데이터 수집, 분석을 신속하게 수행할 수 있다. 데이터 수집과 고속, 대용량 데이터 저장에 적합한 분산 파일시스템과 고속 시계열 기능을 기반으로 의존도, 유사도 분석실행을 제공하는 고속 전처리 시스템의 특징을 제안한다.

• Proceedings of the KIPE Conference
• /
• 2020.08a
• /
• pp.359-360
• /
• 2020

리튬 이온 배터리가 전기 자동차의 주 동력원으로 사용됨에 따라 배터리의 잔존 수명 예측기술의 중요성이 부각되고 있다. 사용 환경에 적합한 잔존 수명 예측을 위해 전기 자동차의 주행 환경을 모사하여 충전 및 방전이 빈번하게 나타나는 UDDS 프로파일에서 범용적으로 사용할 수 있는 수명 인자를 선정하는 것이 필수적이다. 배터리의 잔존 용량과 가장 상관도가 높은 수명 인자를 선정함으로써, 인공지능 기반 예측 알고리즘의 정확도 향상을 기대 할 수 있으며, 태양광 ESS와 같은 상이한 특성의 어플리케이션에도 범용적인 적용이 가능하다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.8
• /
• pp.344-351
• /
• 2016

In the point of view of a train operator, the main concern with a train operation is not only to maintain a time schedule, but also to decrease the energy consumption as much as possible. Generally for a manual drive, a train conductor controls the train acceleration and deceleration by controlling the notches not to exceed the regulation velocity by considering the given maximum velocity profile for an operation route. For this case, the guideline for a conductor is needed to choose the proper notches by applying the notch optimization so as to drive at the regulation velocity and minimize energy consumption simultaneously. In this paper, the real-time notch optimization plan is suggested using a genetic algorithm that optimizes the notches for the remaining route in real time when the event occurs that track information or regulation velocity profile of the remaining route changes during train operation as well as a normal operation situation. An energy saving effect and the convergence behavior of the optimal solution obtained was analyzed in a genetic algorithm.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.35 no.6
• /
• pp.556-561
• /
• 2022

Computational material science as an application of Density Functional Theory (DFT) to the discipline of material science has emerged and applied to the research and development of energy materials and electronic materials such as semiconductor. However, there are a few difficulties, such as generating input files for various types of materials in both the same calculating condition and appropriate parameters, which is essential in comparing results of DFT calculation in the right way. In this tutorial status report, we will introduce how to create crystal structures and to prepare input files automatically for the Vienna Ab initio Simulation Package (VASP) and Gaussian, the most popular DFT calculation programs. We anticipate this tutorial makes DFT calculation easier for the ones who are not experts on DFT programs.

• Science of Emotion and Sensibility
• /
• v.21 no.3
• /
• pp.141-150
• /
• 2018

The purpose of this study is to investigate three-dimensional embroidery techniques for creating conductive fabric materials. Such techniques can increase the efficiency of energy harvesting by increasing the fabric's area during rubbing and brushing. We also investigate the fabric structure of the triboelectric energy harvesting type. Two experiments were conducted for this purpose. In Experiment I, the three-dimensional embroidery technique(satin technique, file technique) and the conductive fabric material(copper-based MPF, nickel-based MPF) were selected as the main variables affecting the efficiency of triboelectric energy harvesting from the human body. Four samples were fabricated according to a combination of two variables. In Experiment II, the harvesters fabricated by the three-dimensional embroidery method showing the highest efficiency were subjected to brushing processes and the voltages generated after processing were analyzed. As a result, in both conductive fabric materials, the pile embroidery fabric structure showed a higher efficiency than the satin structure. These results show the triboelectric energy harvesting principle, which is proportional to the charge density and the generated voltage. It can be seen that the structure of pile embroidery fabric with a large friction area is advantageous for increasing efficiency compared to satin embroidery-fabric structure with a relatively small friction area. Moreover, the energy harvesting efficiency after brushing was higher than that before processing due to the increased friction area, and it was found that the brushing method is advantageous for increasing the triboelectric-energy harvest.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.7
• /
• pp.612-620
• /
• 2013

Problem of energy worldwide reduction had to research on the requirements of services and an important element for the energy management of public buildings and facilities. In this paper, we research the SmartGrid case studies and services requirements based on M2M service. As a result, operation of devices for energy management, network, and application services for a variety of energy-efficiency requirements for the profile was defined. In this study is possible to provide in survey of the smart grid system of M2M-based in public institutions, schools and universities, and large buildings.

• Journal of KIISE:Information Networking
• /
• v.32 no.5
• /
• pp.615-624
• /
• 2005

In Mobile IP, several types of messages - binding update, binding request and binding acknowledgement - are used to support user mobility. It is necessary to exchange those messages frequently for seamless mobility but it incurs both the increase of network overhead and poor usage of mobile node battery power Thus, we need a mechanism that the server detects users location and also copes with the problems effectively, which is our main concern in this paper Each user records all moving logs locally and periodically makes out profile based on them in HMIPv6. By using profile, estimated resident time can be computed whenever he enters an area and the time is set up as the binding update message lifetime. Of course, the more correct lifetime nay be obtained IP arrival time as well as average resident time Is considered in profile. Through extensive experiments, we measure the bandwidth usage for binding update messages by comparing the proposed schemes with that in HMIPv6. From the results, Gain gets over $80\%$ when mobile node stays more than 13 minutes in a subnet. Namely, we come to know that our schemes improve network usage and energy usage in mobile node by decreasing the number of messages while they also manage users locations like that in HMIPv6.

• Radiation Oncology Journal
• /
• v.15 no.1
• /
• pp.65-69
• /
• 1997

Purpose : To obtain the actual dose distribution from measured data by doconvolution method using the measured ion chamber response function. Materials and Methods : The chamber response functions for 2 ionization chambers (diameter 5mm, 6.4mm) were measured. and dose Profiles were measured for $10{\times}20cm^2$ field size using two different detectors. The deconvolution of chamber response function from the measured data were performed for these Profiles. The same procedures were repeated for 4MV, 6MV and 1 SMV photon energies. Results : Different dose Profiles were obtained for the same field with the chambers which have the different response functions. Nearly the same results could be obtained with deconvolution for the profiles from various detectors. Conclusion : The effect of the chamber response function can be extracted by deconvolution method. Deconvolved dose profile using various ionization chambers gave better dose distributions. Technical improvements are needed for practical application.

• Journal of Energy Engineering
• /
• v.6 no.1
• /
• pp.87-95
• /
• 1997

A chemical heat pump based on the reversible reactions between metal chlorides and ammonia gas is attractive alternative to compression system and liquid absorption systems in cooling and refrigerating fields. The advantages of chemical heat pump are no regulatory constants due to CFC refrigerants, utilization of gas, industrial waste heat, electricity, fuel oil etc. as heat sources and wide applications to energy storage system, large-scale energy managements for industrial process. The scale-up of chemical heat pump from laboratory prototype to pilot plants necessitates the interpretation of system performance and evaluation of dynamic behavior in the chemical reactor. This study contains the prediction of performance of chemical refrigerator according to operating condition, the dynamic simulations through reactor modelling, which is used for the calculation of reactive medium temperature and the conversion variation with reactor cooling temperature, and the effect survey of block parameters on the power of refrigerator.