http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Design and analysis of multi-stage expander processes for liquefying natural gas
Wonsub Lim,문일,Inkyu Lee,Kwang-Hee Lee,Byeonggil Lyu,김정환 한국화학공학회 2014 Korean Journal of Chemical Engineering Vol.31 No.9
Multi-stage expander refrigeration cycles were proposed and analyzed in order to develop an efficient naturalgas liquefaction process. The proposed dual and cascade expander processes have high efficiency and the potentialfor larger liquefaction capacity and are suitable for small-scale and offshore natural gas liquefaction systems. Whilerefrigeration cycles of conventional expander processes use pure nitrogen or methane as a refrigerant, the proposedrefrigeration cycles use one or more mixtures as refrigerants. Since mixed refrigerants are used, the efficiency of theproposed multi-stage expander processes becomes higher than that of conventional expander processes. However, theproposed liquefaction processes are different from the single mixed refrigerant (SMR) and dual mixed refrigerant (DMR)processes. The proposed processes use mixed refrigerants as a form of gas, while the SMR and DMR processes usemixed refrigerants as a form of gas, liquid- or two-phase flow. Thus, expanders can be employed instead of Joule-Thomson(J-T) valves for refrigerant expansion. Expanders generate useful work, which is supplied to the compressor, whilethe high-pressure refrigerant is expanded in expanders to reduce its temperature. Various expander refrigeration cyclesare presented to confirm their feasibility and estimate the performance of the proposed process. The specific work,composite curves and exergy analysis data are investigated to evaluate the performance of the proposed processes. Alower specific work was achieved to 1,590 kJ/kg in the dual expander process, and 1,460 kJ/kg in the cascade expanderprocess. In addition, the results of exergy analysis revealed that cycle compressors with associated after-coolers andcompanders are main contributors to total exergy losses in proposed expander processes.
Lee, Junghoon,Kim, Donghyun,Choi, Chang-Hwan,Chung, Wonsub Elsevier 2017 Nano energy Vol.31 No.-
<P><B>Abstract</B></P> <P>Various types of nanoporous anodic aluminum oxide layers and their sealings were studied to enhance the thermal emissivity and hence improve the heat dissipation of aluminum alloy for energy application. Dissipating heat fluxes from the anodized aluminum surfaces were measured using a modified steady-state method and investigated with respect to the various nanoporous morphologies obtained with different anodizing conditions and sealing methods. Results show that the anodized nanoporous oxide layers significantly enhance the thermal emissivity and heat dissipation of aluminum alloy, compared to bare aluminum alloy, and such enhancement is further improved with sealings. A thicker nanoporous oxide layer anodized in oxalic acid results in higher thermal emissivity and better heat dissipation than that in sulfuric acid, showing a darker color which is attributed to the more irregular and disordered pore size and pattern of the nanoporous oxide layer. The nanoporous oxide layer with cold NiF<SUB>2</SUB> or black sealing shows further enhancement in thermal emissivity and heat dissipation, demonstrating the highest enhancement in emissivity up to 0.906 in case of the nanoporous oxide layer anodized in oxalic acid with black sealing, which is seven times greater than that of bare aluminum. The nanoporous oxide layer with black sealing also results in the significant improvement of the cooling efficiency of a heat exchanger system of aluminum alloy by 36.4%, suggesting great energy saving for real energy application.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Porous anodic alumina nanostructures enhance radiative heat transfer of aluminum. </LI> <LI> The sealing of porous nanostructures further enhances the thermal energy transfer. </LI> <LI> Enhancement of the thermal emissivity up to 700% and the heat flux up to 300%. </LI> <LI> Demonstration of the enhancement of heat dissipation for a real heat exchange system. </LI> </UL> </P> <P><B>Graphical abstract</B></P> <P>[DISPLAY OMISSION]</P>
Black Pine Bast Scale Detection using Deep Learning
( Wonsub Yun ),( J. Praveen Kumar ),( Sangjoon Lee ),( Byoung-kwan Cho ) 한국농업기계학회 2021 한국농업기계학회 학술발표논문집 Vol.26 No.1
Pests play a major role in depletion of the agricultural resources. The pest detection can help to prevent the depletion. Even though there is a huge development in technologies, the current farm management devices and methods do not meet the required level to detect the precise pests in the farms and forests. Moreover, in practice, most of the pest detection methods are conventional in nature and depend on the professional workers. The drawbacks of the conventional methods are high cost, time consuming, knowledge dependency of the professional workers, etc. To overcome these drawbacks, a deep learning based pest detection model has been developed to detect the pest. In our research, the key focus is to detect the black pine bast scale. The pheromone traps are used to lure these pests. In order to capture the pheromone trap images effectively, an image capture set up has been developed. It is helpful in solving the problems such as non-uniform image capturing distance and the reflection caused by the outer vinyl present in the pheromone trap. In this experiment, the smartphone Huawei P30 Pro model is used to capture the images. In order to obtain better result from the captured smartphone images, the images were cropped into image segments. These image segments are given as input for the deep learning models. The pre-trained models used in this experiment are Fast-RCNN, Faster-RCNN, and RetinaNet. The ResNet50, ResNet101, and ResNext101 are used as the backbone layers. Among these developed model combinations, RetinaNet ResNet101 with a combination of the FPN as the backbone layer attains the highest F1 score of 0.78. Hence, this model can be used for automatic detection of black pine bast scale pests attached to the pheromone traps. Then, an image stitching algorithm is used to merge the image segments. Finally, a smartphone application is developed for Black Pine Bast Scale detection.
Institutional Complementarity during Policy Emulation
Wonsub Eum(엄원섭),Jeong-Dong Lee(이정동) 한국산업경제학회 2015 한국산업경제학회 정기학술발표대회 논문집 Vol.2015 No.12
Importance of complementarity among institutional factors in the policy system has been highlighted recently. This paper tries to specifically examine institutional complementarity during policy emulation, by looking at the international cases of general trading companies in 1970 to 80s, from Korea, Taiwan, and United States. The results argue that even with a successfully proven policy would need to consider the relevant institutional backgrounds of a country during adoption, and the different results from similar policy in different countries come from the difference in institutional complementarity in their formal and informal rules.
14자유도 전차량 모델 기반 섀시제어시스템 Dynamic Sensor 고장 검출 알고리즘 개발
최원섭(Wonsub Choi),이재훈(Jaehoon Lee),국형석(Hyung-Seok Kook),허승진(Seung-Jin Heo) 한국자동차공학회 2010 한국자동차공학회 학술대회 및 전시회 Vol.2010 No.11
With the advent of the new technologies in electronics, vehicle dynamics, and sensors, intelligent chassis control systems have been widely adopted in vehicles, recently. These chassis control systems use various sensors to measure real-time signals from the operating vehicle on roads such as steering angle, wheel speeds, vehicle’s yaw rate and lateral acceleration. Therefore, an unexpected sensor fault in the chassis control system could lead to serious damage to the vehicle, and in serious cases could be threatening to the human safety. In the present article, a reliable sensor fault detection method is concerned. The sensor fault detection algorithm proposed in the present work is based on a full vehicle model developed in a previous work. The high fidelity modeling capability using analytical and empirical models based on 14 degree of freedom model and the real-time computation capability of the full vehicle model enable computed virtual sensor signals to be compared with the real sensor signals and even be substituted for in case of sensor fault. Detailed algorithm for the sensor fault detection is described, and simulated results for the cases of lateral acceleration sensor and yaw rate sensor faults are demonstrated in the present article.
( Sanghyuk Lee ),( Donghyun Kim ),( Kyungsik Son ),( Yongje Choi ),( Wonsub Chung ) 대한금속재료학회(구 대한금속학회) 2016 대한금속·재료학회지 Vol.54 No.5
A typical brightener-propargyl alcohol-was added to enhance the cutting performance during Ni-diamond composite electrodeposition. Electrochemical analysis was performed and mechanical properties such as hardness and wear resistance were examined. In addition, the surface morphology of composite coating layers was observed using an optical microscope, and using image analysis software, the dispersivity of the diamond particles was analyzed to calculate the number of single diamond particles. A galvanostatic test was employed to identify the electrodeposition mechanism as a function of the concentration of propargyl alcohol. When 0.1 mg/ℓ of propargyl alcohol was added, the dispersivity and adhesion performances of the Ni-diamond coating layer were optimized for cutting tools. (Received September 22, 2015)