• Computers and Concrete
• /
• 제16권5호
• /
• pp.741-757
• /
• 2015

In the last decade, several modeling approaches have been proposed and applied to estimate the high-performance concrete (HPC) slump flow. While HPC is a highly complex material, modeling its behavior is a very difficult issue. Thus, the selection and application of proper modeling methods remain therefore a crucial task. Like many other applications, HPC slump flow prediction suffers from noise which negatively affects the prediction accuracy and increases the variance. In the recent years, ensemble learning methods have introduced to optimize the prediction accuracy and reduce the prediction error. This study investigates the potential usage of bagging (Bag), which is among the most popular ensemble learning methods, in building ensemble models. Four well-known artificial intelligence models (i.e., classification and regression trees CART, support vector machines SVM, multilayer perceptron MLP and radial basis function neural networks RBF) are deployed as base learner. As a result of this study, bagging ensemble models (i.e., Bag-SVM, Bag-RT, Bag-MLP and Bag-RBF) are found superior to their base learners (i.e., SVM, CART, MLP and RBF) and bagging could noticeable optimize prediction accuracy and reduce the prediction error of proposed predictive models.

• 한국재료학회지
• /
• 제20권2호
• /
• pp.90-96
• /
• 2010

The recirculating electrochemical flow reactor developed at UCLA has been employed to fabricate nanostructured GMR multilayers. For comparison, Ni/Cu multilayers have been electrodeposited from a single bath, from dual baths and from the recirculating electrochemical flow reactor. For a magnetic field of 1.5 kOe, higher GMR (Max. -5%) Ni/Cu multilayers with low electrical resistivity (< $10\;{\mu}{\Omega}{\cdot}cm$) were achieved by the electrochemical flow reactor system than by the dual bath (Max. GMR = -4.2% and < $20\;{\mu}{\Omega}{\cdot}cm$) or the single bath (Max. GMR = -2.1% and < $90\;{\mu}{\Omega}{\cdot}cm$) techniques. Higher GMR effects have been obtained by producing smoother, contiguous layers at lower current densities and by the elimination of oxide film formation by conducting deposition under an inert gas environment. Our preliminary GMR measurements of Ni/Cu multilayers from the electrochemical flow reactor obtained at low magnetic field of 0.15 T, which may approach or exceed the highest reported results (-7% GMR) at magnetic fields > 5 kOe.

• 센서학회지
• /
• 제15권1호
• /
• pp.47-52
• /
• 2006

This paper describes the design, fabrication and characteristics of a buckling microvalve using a MCA (multilayer ceramic actuator). The mechanical and fluidic analysis are done by finite element method. The designed structure is normally closed microvalve using buckling effect, which is consist of three separate structures; a valve seat die, an actuator die and a small piezoelectric actuator. The flow rate of the fabricated MCA valve was 0-8.13 ml/min at the applied pressure of 0-50 kPa. Maximum non-linearity was 2.24 % FS at a duty cycle of 50 %. The maximum pressure was 230 kPa and the leak rate was $3.03{\times}10^{-8}\;Pa{\cdot}m^{3}/cm^{2}$ at a supply voltage of 100 V.

• 한국재료학회지
• /
• 제20권4호
• /
• pp.187-193
• /
• 2010

Chromium nitride (CrN) samples with two different layer structures (multilayer and single layer) were coated on bipolar plates of polymer electrolyte membrane fuel cells (PEMFC) using the reactive sputtering method. The effects with respect to layer structure on corrosion resistance and overall cell performance were investigated. A continuous and thin chromium nitride layer ($Cr_{0.48}\;N_{0.52}$) was formed on the surface of the SUS 316L when the nitrogen flow rate was 10 sccm. The electrochemical stability of the coated layers was examined using the potentiodynamic and potentiostatic methods in the simulated corrosive circumstances of the PEMFC under $80^{\circ}C$. Interfacial contact resistance (ICR) between the CrN coated sample and the gas diffusion layer was measured by using Wang's method. A single cell performance test was also conducted. The test results showed that CrN coated SUS316L with multilayer structure had excellent corrosion resistance compared to single layer structures and single cell performance results with $25\;cm^2$ in effective area also showed the same tendency. The difference of the electrochemical properties between the single and multilayer samples was attributed to the Cr interlayer layer, which improved the corrosion resistance. Because the coating layer was damaged by pinholes, the Cr layer prevented the penetration of corrosive media into the substrate. Therefore, the CrN with a multilayer structure is an effective coating method to increase the corrosion resistance and to decrease the ICR for metallic bipolar plates in PEMFC.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2011년도 제41회 하계 정기 학술대회 초록집
• /
• pp.127-127
• /
• 2011

We investigate the transparent TFTs using a transparent ZnSnO3 (ZTO)/Ag/ZTO multilayer electrode as S/D electrodes with low resistivity of $3.24{\times}10^{-5}$ ohm-cm, and high transparency of 86.29% in ZTO based TFTs. The Transparent TFTs (TTFTs) are prepared on glass substrate coated 100 nm of ITO thin film. On atomic layer deposited $Al_2\;O_3$, 50 nm ZTO layer is deposited by RF magnetron sputtering through a shadow mask for channel layer using ZTO target with 1 : 1 molar ratio of ZnO : $SnO_2$. The power of 100W, the working pressure of 2mTorr, and the gas flow of Ar 20 sccm during the ZTO deposition. After channel layer deposition, a ZTO (35 nm)/Ag (12 nm)/ZTO(35 nm) multilayer is deposited by DC/RF magnetron sputtering to form transparent S/D electrodes which are patterned through the shadow mask. Devices are annealed in air at 300$^{\circ}C$ for 30 min following ZTO deposition. Using UV/Visible spectrometer, the optical transmittances of the TTFT using ZTO/Ag/ ZTO multilayer electrodes are compared with TFT using Mo electrode. The structural properties of ZTO based TTFT with ZTO/Ag/ZTO multilayer electrodes are analyzed by high resolution transmission electron microscopy (HREM) and X-ray photoelectron spectroscopy (XPS). The transfer and output characterization of ZTO TTFTs are examined by a customized probe station with HP4145B system in are.

• Advances in aircraft and spacecraft science
• /
• 제1권1호
• /
• pp.27-41
• /
• 2014

An active flow control technique based on "smart-tabs" is proposed to delay flow separation on a circular cylinder. The actuators are retractable and orientable multilayer piezoelectric tabs which protrude perpendicularly from the model surface. They are mounted along the spanwise direction with constant spacing. The effectiveness of the control was tested in pre-critical and in post-critical regime by evaluating the effects of several control parameters of the tabs like frequency, amplitude, height, angular position and plate incidence with respect to the local flow. Measurements of the mean static pressure distribution around the cylinder were used to estimate the pressure drag coefficient. The maximum drag reduction achieved in the pre-critical regime was of the order of 30%, whereas in the post-critical regime was about 10%, 3% of which due to active forcing. Furthermore, pressure fluctuation measurements were performed and spectral analysis indicated an almost complete suppression of the vortex shedding in active forcing conditions.

• 한국광학회지
• /
• 제15권4호
• /
• pp.361-368
• /
• 2004

콤팩트디스크(CD-R; Compact DiskRecordable)를 성분별로 분리하여 제작하고, 다층 박막 구조를 만들어서 레이저빔의 에너지를 변화시켜 가면서 조사하여 각 성분 층에서의 최적 미소 점 마킹 조건과 홈 형성 과정에 관하여 연구하였다. 본 연구는 Q-스위치 Nd:YAG 레이저를 이용하여 준비된 각 시료의 표면에 27∼373 mJ 빔을 80 $\mu\textrm{m}$의 점적 크기로 조사하여 샘플에 형성된 흠 형태를 광학현미경(OM; Optical Microscope)과 광 결맞음 단층촬영기(OCT; Optical Coherence Tomography)로 비교-관찰하여 미소 점 마킹의 형성 과정을 분석하였다. 다층 박막에서 용융된 기판 층은 짧은 시간동안 충분한 열 흐름이 발생하여 증배의 형성을 야기하며, 반사 층과 염료 층 사이에 흡수된 에너지는 염료를 용융시키고 체적을 증가시켰으며, 증배가 식으면서 표면장력의 영향 및 레이저빔에 의한 순간적인 시편의 온도상승으로 기화와 반동압력에 의한 질량흐름 때문에 두 층의 경계면에서 홈과 외륜의 발생을 설명할 수 있었다. 따라서 다층 박막에서의 미소 점 마킹의 형성 과정은 표면장력, 용융 점성력, 층 두께, 다층 박막 성분 물질의 물리화학적ㆍ광학적 성질과 관계가 있음을 알 수 있었다.

• 한국통신학회논문지
• /
• 제33권8B호
• /
• pp.707-718
• /
• 2008

Traffic classification의 방법은 동적으로 변하는 application의 변화에 대처하기 위하여 페이로드나 port를 기반으로 하는 것에서 ML 알고리즘을 기반으로 하는 것으로 변하여 가고 있다. 그러나 현재의 ML 알고리즘을 이용한 traffic classification 연구는 offline 환경에 맞추어 진행되고 있다. 특히, 현재의 기존 연구들은 testing 방법으로 cross validation을 이용하여 traffic classification을 수행하고 있으며, traffic flow를 기반으로 classification 결과를 제시하고 있다. 본 논문에서는 testing방법으로 cross validation과 split validation을 이용했을 때, traffic classification의 정확도 결과를 비교한다. 또한 바이트를 기반으로 한 classification의 결과와 flow를 기반으로 한 classification의 결과를 비교해 본다. 본 논문에서는 J48, REPTree, RBFNetwork, Multilayer perceptron, BayesNet, NaiveBayes와 같은 ML 알고리즘과 다양한 feature set을 이용하여 트래픽을 분류한다. 그리고 split validation을 이용한 traffic classification에 적합한 최적의 ML 알고리즘과 feature set을 제시한다.

• Computers and Concrete
• /
• 제25권6호
• /
• pp.565-574
• /
• 2020

Over the years, several machine learning approaches have been proposed and utilized to create a prediction model for the high-performance concrete (HPC) slump flow. Despite HPC is a highly complex material, predicting its pattern is a rather ambitious process. Hence, choosing and applying the correct method remain a crucial task. Like some other problems, prediction of HPC slump flow suffers from abnormal attributes which might both have an influence on prediction accuracy and increases variance. In recent years, different studies are proposed to optimize the prediction accuracy for HPC slump flow. However, more state-of-the-art regression algorithms can be implemented to create a better model. This study focuses on several methods with different mathematical backgrounds to get the best possible results. Four well-known algorithms Support Vector Regression, M5P Trees, Random Forest, and MLPReg are implemented with optimum parameters as base learners. Also, redundant features are examined to better understand both how ingredients influence on prediction models and whether possible to achieve acceptable results with a few components. Based on the findings, the MLPReg algorithm with optimum parameters gives better results than others in terms of commonly used statistical error evaluation metrics. Besides, chosen algorithms can give rather accurate results using just a few attributes of a slump flow dataset.

• 대한전자공학회논문지
• /
• 제26권6호
• /
• pp.48-56
• /
• 1989

밀착형 1차원 영상감지소자로서 팩시밀리에 사용될 광도전막을 사일랜듸 글로방전 분해법으로 제작하였다. 우선 rf전력, 사일랜유량, 분위기 가스압, $H_2/SiH_4$비 및 기판온도의 증착조건에 따른 단층광전도막의 전기적 및 광학적특성을 조사하였다. 이 단층구조 영상감지막은 광전감도 0.85와 100lux 조도하에서 $I_{ph}/I_d=100$을 나타내었다. 그러나 이러한 단층박막은 양 전극으로 부터의 캐리어주입으로 인해 큰 암전류도 0.2nA 이하를 나타내었다. 또한 다층막은 단층막에 비해 단파장 가시광영역이 보상되어 팩시밀리용 1차원 영상감지소자에 사용될 만한 결과를 나타내었다.