PURPOSES : Despite the availability of larger traffic data and more advanced data collection methods, the problem of missing data is yet to be solved. Imputing missing data to ensure data quality and reliability of statistics has always been challenging. Missing data are imputed via several existing methods, such as autoregressive integrated moving average, exponential smoothing, and interpolation. However, these methods are complicated and results in significant errors. METHODS : A deep-learning method was applied in this study to impute traffic volume data of the South Korean national highway. Traffic data were trained using the long short-term memory method, which is a suitable deep-learning method for time series analysis. RESULTS : Three cases were proposed to estimate the traffic volume. In the first case, which represented the general conditions, the mean absolute percentage error (MAPE) was 12.7%. The second estimation case, which was based on the opposite traffic flow, exhibited a MAPE of 17%~18%. The third case, which was estimated using adjacent-section data, had a MAPE of 18.2%. CONCLUSIONS : Deep learning may be a suitable alternative data imputation method based on the limited site and data. However, its application depends on the specific situation. Furthermore, deep-learning models can be improved using an ensemble method, batch-size, or through model-structure optimization.

In this paper, graphene-coated Al powders prepared by in situ reduction method were directly used for cold spraying, obtaining a graphene-reinforced Al matrix composite coating with more compact structure and better performance. Cross-sectional analysis revealed that compared with the pure Al powders, the graphene-coated Al powders were more severely deformed, and the resulting coating was denser and its porosity was reduced by over 80%. The hardness of the graphene-coated Al coating was increased by 40%, and its brine immersion time was prolonged by nearly three times. However, the graphene increases the pitting sensitivity of the Al coating; so, the enhanced corrosion resistance of the graphene-coated Al coating is mainly attributed to the improvement of its structure densification.

Although flame synthesis promises economic benefit and rapid synthesis of carbon nanotube (CNT), the lack of control and understanding of the effects of flame parameters (e.g., temperature and precursor composition) impose some challenges in modelling and identifying CNT growth region for obtaining better throughput. The present study presents an investigation on the types of carbon precursor that affect CNT growth region on nickel catalyst particles in an ethylene inverse diffusion flame. An established CNT growth rate model that describes physical growth of CNT is utilised to predict CNT length and growth region using empirical inputs of flame temperature and species composition from the literature. Two variations of the model are employed to determine the dominant precursor for CNT growth which are the constant adsorption activation energy (CAAE) model and the varying adsorption activation energy (VAAE) model. The carbon precursors investigated include ethylene, acetylene, and carbon monoxide as base precursors and all possible combinations of the base precursors. In the CAAE model, the activation energy for adsorption of carbon precursor species on catalyst surface E a,1 is held constant whereas in the VAAE model, E a,1 is varied based on the investigated precursor. The sensitivity of the growth rate model is demonstrated by comparing the shifting of predicted growth regions between the CAAE model and the VAAE model where the CAAE model serves as a control case. Midpoint-based and threshold-based techniques are employed within each model to quantify the predicted CNT growth region. Growth region prediction based on the midpoint-VAAE approach demonstrates the importance of acetylene and carbon monoxide to some extent towards CNT growth. Ultimately, the threshold-VAAE model shows that the dominant precursor for CNT growth is the mixture of acetylene and carbon monoxide. A simplified reaction mechanism is proposed to describe the surface chemistry for precursor reactions with nickel catalyst where decomposition of the ethylene fuel source into acetylene and carbon monoxide is accounted for by chemisorption.

This work reports the syntheses of an inexpensive and efficient asphalt-derived mesoporous carbon (AdMC) as an adsorbent. The adsorbent was activated with potassium hydroxide to increase its surface area and then characterized by SEM–EDS, FT-IR, and BET. The adsorption properties of AdMC were evaluated for the adsorptive removal of eleven Poly Aromatic Hydrocarbons (PAHs) and diesel from water samples. The prepared AdMC showed very high surface areas and high micropore volumes equal to 2316 m2/g and 1.2 cm3/g, respectively. Various experimental conditions influencing the adsorption capacity of eleven PAHs and diesel were investigated. At high concentrations, PAHs and diesel solubility in water is very low. Hence, samples were emulsified with a surfactant, and then maximum adsorption capacity was investigated. Adsorption profile of individual PAHs was examined using gas chromatography/mass spectrometry analysis followed by liquid–liquid extraction. Total hydrocarbon removal was studied using a total organic analyzer. Asphalt-derived mesoporous sorbent showed an extreme ability to remove PAHs and diesel (average adsorption capacity of 166 mg/g for individual PAHs and diesel (maximum capacity of 1600 mg/g). The experimental results fitted the Langmuir model with a correlation efficiency of 0.9853. The results obtained for both adsorbents also matched to pseudo-second-order kinetics, suggesting that the adsorption of PAHs and diesel is chemical, monolayer, and homogeneous process.

The present work focused on the determination of texture, morphology, crystallinity, and gas adsorption characteristics of porous graphene prepared from rice husks ashes at different stabilization temperature. The stabilization temperature applied in this work is 100 °C, 200 °C, 300 °C, and 400 °C to convert rice husk into rice husk ashes (RHA). Chemical activation was adopted at temperature 800 °C using potassium hydroxide (KOH) as dehydrating agent at (1:5) impregnation ratio to convert RHA into rice husk ashes-derived graphene (GRHA). The resultant GRHA were characterized in terms of their morphological changes, SSA, crystallinity, and functional group with TEM, the BET method, Raman spectroscopy, and XRD analysis, respectively. Results from this study showed that the SSA of the GRHA at stabilization temperature 200 °C (1556.3 m2/g) is the highest compared to the other stabilization temperature. Raman spectroscopy analysis revealed that all GRHA samples possess D, G, and 2D bands, which confirm the successful synthesis of the rice husks into porous graphene-like materials, known as GRHA. Appearance of diffraction peak in XRD at 44.7° indicating the graphitic structure of all the GRHA samples. Meanwhile, the TEM images of GRHA200 exhibited wrinkled structures due to the intercalation of oxygen and a few layers of graphene flakes. These wrinkled structures and graphene layers are the other factors that lead to the highest SSA of GRHA200 compared to other prepared samples GRHA. Furthermore, the adsorption capacity of CH4 for GRHA200 is up to 43 cm3/g at 35 bar and ambient temperature, almost double the adsorption capacity performance of GRHA400 at the same operating pressure and temperature.

The aim of this work is to investigate the ability of a new functionalized graphene oxide 3-amino-5-phenylpyrazole (F-GO) in the adsorption and removal of Hg2+ from aqueous solution. Both untreated graphene oxide (GO) and F-GO were characterized using FT-IR, EDX, FE-SEM, XRD and TGA analysis. The effects of three operational variables (pH, adsorbent dose and initial metal ion concentrations) on Hg2+ adsorption capacity of F-GO were investigated by central composite design. This technique aims to find a simple way to optimize the adsorption process and to analyze the interaction between the significant parameters. A quadratic model suggested for the analysis of variance found that the adsorption of metal ions heavily depend upon pH of the solution. The adsorption mechanism has been determined by pseudo-first-order kinetic models and the adsorption behavior was modeled by Freundlich isotherm. Results demonstrated that the adsorption capacities of F-GO for removal of Hg2+ were generally higher than those of GO, which is attributed to a decrease in the agglomeration of graphene layers due to the presence of amino-functional moieties with their bulky phenyl groups. Thermodynamic data indicated that the functionalization significantly affects the thermostability of the GO precursor materials. The desorption study demonstrated favorable regenerability of the F-GO adsorbent, even after three adsorption–desorption cycles.

Giant miscanthus (GM) is an Asian grass that can produce biomass in high yields per land area. It can be used as a cathode material in lithium sulfur (Li/S) batteries. Giant-miscanthus-derived activated carbon (GMAC) is prepared via carbonization of GM followed by KOH activation. It is prepared with a large amount of KOH, and thus contained more defects but had a highly porous structure and graphitic cluster lattice. GMAC has a large specific surface area of 3327 m2/g and a large total pore volume of 1.86 cm3/g. The pore volume served as a storage space for the retention of polysulfides, thereby inhibiting the shuttle effect. When a GMAC–sulfur composite cathode is tested in a Li/S battery, an initial discharge capacity of 1148 mAh/g can be attained at 0.1 C. In a cyclic charge–discharge experiment at 1 C, discharge capacities of 529 mAh/g and 248 mAh/g are observed in the first and 200th cycles, respectively.

홍콩에는 예부터 지금까지 파키스탄, 인도, 네팔 등에서 온 이민들을 포함하여, 비중국어권 아동들이 적지 않다. 평등교육의 이념을 표방하는 홍콩평등기회위원회(香港平等機會委員會)는 비중국어권 학습자들이 중국어권 학습자들과 같은 학교에서 학습하기를 건의하여 그들이 최대한 빠른 시일 내에 홍콩 사회로 융화되게 하려 하였다. 이로 인해, 비중국어권 아동들은 홍콩 현지의 일반 유치원을 다니게 되었는데, 홍콩의 교육 관할 당국도 적지 않은 자원을 투입하여 이들이 현지 교육과정에 적응할 수 있도록 협조하였으나, 그 효과를 즉각적으로 확인할 수는 없었다. 그 중 가장 큰 문제는 학습의 차이를 처리하는 것이었다. 비중국어권 아동들의 중국어 실력은 모국어 수준에 미치지는 못했다. 이들이 중국어권 아동들과 함께 중국어 수업에 참여할 때 적잖은 어려움이 발행했다. 또한, 비중국어권 아동들의 성장 배경도 각기 달랐다. 신이민자 가정의 아동도 있고, 현지에서 출생한 소수 민족 아동도 있어, 이러한 학습자 사이의 중국어 수준의 차이도 존재했다. 이밖에도 소수민족과 홍콩의 문화적 차이 또한 매우 커서 모든 교사와 아동은 ‘학습’과 ‘교수’ 두 측면에서 모두 여러 가지 차이로 인해 발생하는 문제들을 처리해야 했다. 이처럼 비중국어권 학습자들이 홍콩에서 중국어를 배우는 것은 매우 쉽지 않았다. 상술한 문제들을 해결하기 위해서 경마회 자선기금 및 다섯 단체의 공영교육 사회복지기구는 ｢경마회의 재밌는 중국어 배우기(賽馬會(有)友趣學中文)｣ 라는 프로젝트를 추진하여, 중국어를 모국어로 갖지 않는 학습자들이 효과적으로 중국어 학습 하여 빠르게 사회에 융화되도록 도왔다. 본 프로 젝트는 수준별 교육과정, 수준별 교재, 수준별 학습법, 수준별 교구, 수준별 평가와 다층적 교학법을 설계하고 지원하여 학습 차이에서 생기는 어려움을 해결하는 데 도움을 주었다. 본고에서는 연구 이론과 설계, 데이터 수집 및 연구 결과를 상세히 서술하고, 효과적인 사례에 대해 상세하게 서술하려고 한다. 이러한 연구 성과는 세계 각지에 광범위하게 적용될 수 있을 것이다.