• Journal of Wetlands Research
• /
• v.25 no.4
• /
• pp.306-314
• /
• 2023

Stream sediments are an important component of water quality management because they are receptors of various pollutants such as heavy metals and organic matters emitted from upland sources and can be secondary pollution sources, adversely affecting water environment. To effectively manage the stream sediments, identification of primary sources of sediment contamination and source-associated control strategies will be required. We evaluated the performance of machine learning models in identifying primary sources of sediment contamination based on the physico-chemical properties of stream sediments. A total of 356 stream sediment data sets of 18 quality parameters including 10 heavy metal species(Cd, Cu, Pb, Ni, As, Zn, Cr, Hg, Li, and Al), 3 soil parameters(clay, silt, and sand fractions), and 5 water quality parameters(water content, loss on ignition, total organic carbon, total nitrogen, and total phosphorous) were collected near abandoned metal mines and industrial complexes across the four major river basins in Korea. Two machine learning algorithms, linear discriminant analysis (LDA) and support vector machine (SVM) classifiers were used to classify the sediments into four cases of different combinations of the sampling period and locations (i.e., mine in dry season, mine in wet season, industrial complex in dry season, and　industrial complex in wet season). Both models showed good performance in the classification, with SVM outperformed LDA; the accuracy values of LDA and SVM were 79.5% and 88.1%, respectively. An SVM ensemble model was used for multi-label classification of the multiple contamination sources inlcuding landuses in the upland areas within 1 km radius from the sampling sites. The results showed that the multi-label classifier was comparable performance with sinlgle-label SVM in classifying mines and industrial complexes, but was less accurate in classifying dominant land uses (50~60%). The poor performance of the multi-label SVM is likely due to the overfitting caused by small data sets compared to the complexity of the model. A larger data set might increase the performance of the machine learning models in identifying contamination sources.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.27 no.3
• /
• pp.239-249
• /
• 2001

The purpose of this study is to evaluate the relationship of the factors which could be influenced by orthognathic surgery especillay SSRO. We measured the amounts of the maximum opening, lateral movements, maximum velocity and pattern of mandibular path during the opening and closing of mandible at the following times ; preoperative, 1 month after operation, 6 months after operation respectively using MKG. And the results were compared according to the categorized subgroups. Following results were obtained : 1. The change of the amounts of mandibular lateral movement and maximum opening velocity were statistically different between male and female (p<0.05), but the others were not. 2. According to the method of operation, there was no difference in the change of the mandibular movements between the group of SSRO and SSRO plus LeFort I osteotomy (p>0.05). 3. According to the amounts of mandibular movement, the recovery of left lateral movement of the group of $6{\sim}10mm$ was better than the other groups (p<0.05). 4. In the frontal pattern of the opening and closing of the mandible, the complex deflected type (F5), simple deflected type (F4), complex deviated type (F3), simple deviated type (F2), straight type (F1) were obtained in order at the time of preoperative, simple deflected type, simple deviated type, complex deviated type, straight type, complex deflected type in order at the time of 1 month after surgery, and the result at the time of 6 months after surgery was the same with that of the time of preoperative. In the sagittal pattern, non-coincident type (S2) was predominant at the time of preoperative, and coincident type (S1) was predominant at the time of 1 month after surgery. After 6 months, the result was also the same with that of the preoperative in sagittal pattern. 5. There was not a statistical difference in the change of the mandibular movement between group of presence of the preoperative TMJ symptoms and non-presence group (p>0.05). 6. There was not a statistical difference in the change of the mandibular movement between repositioning device applied group and non-applied group (p>0.05). 7. Sixty three percents of the patients who had preoperative TMJ symptoms were improved after surgery and preoperative TMJ symptoms were more improved after operation in the repositioning device non-applied group statistically (p<0.05).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.12
• /
• pp.2899-2904
• /
• 2015

This paper introduces about the effect on $I_{DS}-V_{GS}$ characteristic of transistor that interface trap charge is created by damage due to heat in a 3D sequential inverter. A interface trap charge distribution in oxide layer in a 3D sequential inverter is extracted using two-dimensional device simulator. The variation of threshold voltage of top transistor according to the gate voltage variation of bottom transistor is also described in terms of Inter Layer Dielectric (ILD) length of 3D sequential inverter, considering the extracted interface trap charge distribution. The extracted interface trap density distribution shows that the bottom $HfO_2$ layer and both the bottom and top $SiO_2$ layer were relatively more affected by heat than the top $HfO_2$ layer with latest process. The threshold voltage variations of the shorter length of ILD in 3D sequential inverter under 50nm is higher than those over 50nm. The $V_{th}$ variation considering the interface trap charge distribution changes less than that excluding it.

• Journal of Microbiology and Biotechnology
• /
• v.21 no.10
• /
• pp.1012-1020
• /
• 2011

A gene coding for an endoglucanase (EglA), of the glycosyl hydrolase family 12 and derived from Aspergillus niger VTCC-F021, was cloned and sequenced. The cDNA sequence, 717 bp, and its putative endoglucanase, a 238 aa protein with a predicted molecular mass of 26 kDa and a pI of 4.35, exhibited 98.3-98.7% and 98.3-98.6% identities, respectively, with cDNA sequences and their corresponding endoglucanases from Aspergillus niger strains from the GenBank. The cDNA was overexpressed in Pichia pastoris GS115 under the control of an AOX1 promoter with a level of 1.59 U/ml culture supernatant, after 72 h of growth in a YP medium induced with 1% (v/v) of methanol. The molecular mass of the purified EglA, determined by SDS-PAGE, was 33 kDa, with a specific activity of 100.16 and 19.91 U/mg toward 1% (w/v) of ${\beta}$-glucan and CMC, respectively. Optimal enzymatic activity was noted at a temperature of $55^{\circ}C$ and a pH of 5. The recombinant EglA (rEglA) was stable over a temperature range of $30-37^{\circ}C$ and at pH range of 3.5-4.5. Metal ions, detergents, and solvents tested indicated a slightly inhibitory effect on rEglA activity. Kinetic constants ($K_m$, $V_{max}$, $k_{cat}$, and $k_{cat}/K_m$) determined for rEglA with ${\beta}$-glucan as a substrate were 4.04 mg/ml, 102.04 U/mg, 2,040.82 $min^{-1}$, and 505.05, whereas they were 10.17 mg/ml, 28.99 U/mg, 571.71 $min^{-1}$, and 57.01 with CMC as a substrate, respectively. The results thus indicate that the rEglA obtained in this study is highly specific toward ${\beta}$-glucan. The biochemical properties of rEglA make it highly valuable for downstream biotechnological applications, including potential use as a feed enzyme.

• Journal of the Korean Geotechnical Society
• /
• v.29 no.6
• /
• pp.5-17
• /
• 2013

Slug tests can be adopted to estimate hydraulic conductivity of the slurry trench wall backfill for its abilities to reflect the in-situ performance of the construction. A comprehensive three-dimensional numerical model is developed to simulate the slug test in a slurry trench wall considering the presence of bentonite cake on the interface boundaries between the wall and the surrounding soil formation. Influential factors such as wall width (i.e., proximity of wall boundary), well deviation, vertical position of well intake section, compressibility of wall backfill, etc. are taken into account in the model. A series of simulation results are examined to evaluate the bentonite cake effect in analyzing practical slug test results in the slurry trench wall. The results show that the modified line-fitting method can be used without any correction factor for the slug test in the slurry trench wall with the presence of bentonite cake. A case study is reanalyzed with the assumption of existing bentonite cake. The results are compared with the previously reported results by the approaches assuming no bentonite cake (constant-head boundary) or upper-bound solution (no-flux boundary). The case study demonstrates the bentonite cake effect and the validity of the modified line-fitting method in the estimation of the hydraulic conductivity of the slurry wall backfill.

• Journal of the Korean Geotechnical Society
• /
• v.28 no.11
• /
• pp.17-31
• /
• 2012

No analytical solution exists for evaluating in-situ hydraulic conductivity of vertical cutoff walls by analyzing slug test results with consideration of transient flow. There is an analytical solution proposed to interpret a slug test performed in a partially penetrated well within an aquifer. However, this analytical solution cannot be directly applied to the cutoff wall because the solution has been developed exclusively for an infinite aquifer instead of a narrow cutoff wall. To consider the cutoff wall boundary conditions (i.e, constant head boundary and no flux boundary condition), the analytical solution has been modified in this study to take into account the narrow boundaries by introducing the imaginary well theory. Type curves are constructed from the currently derived analytical solution and compared with those of a partially penetrated well within an aquifer. The constant head boundary condition provides faster hydraulic head recovery curve than the aquifer case. On the other hand, no flux boundary condition leads to slower hydraulic head recovery. The bigger the shape factor and deviation of the well and the smaller the width of the vertical cutoff wall are, the more effect of boundary condition was observed. The type curves obtained from the analytical solution for a cutoff wall are similar to those made by the numerical method in the literature.

• Bulletin of the Korean Chemical Society
• /
• v.2 no.3
• /
• pp.96-104
• /
• 1981

(1) The flow data of f (stress) and ${\dot{s}$ (strain rate) for Fe and Ti alloys were plotted in the form of f vs. -ln ${\dot{s}$ by using the literature values. (2) The plot showed two distinct patterns A and B; Pattern A is a straight line with a negative slope, and Pattern B is a curve of concave upward. (3) According to Kim and Ree's generalized theory of plastic deformation, pattern A & B belong to Case 1 and 2, respectively; in Case 1, only one kind of flow units acts in the deformation, and in Case 2, two kinds flow units act, and stress is expressed by $f={X_1f_1}+{X_2f_2}$where $f_1\;and\;f_2$ are the stresses acting on the flow units of kind 1 and 2, respectively, and $X_1,\;X_2$ are the fractions of the surface area occupied by the two kinds of flow units; $f_j=(1/{\alpha}_j) sinh^{-1}\;{\beta}_j{{\dot{s}}\;(j=1\;or\;2)$, where $1/{\alpha}_j\;and\;{\beta}_j$ are proportional to the shear modulus and relaxation time, respectively. (4) We found that grain-boundary flow units only act in the deformation of Fe and Ti alloys whereas dislocation flow units do not show any appreciable contribution. (5) The deformations of Fe and Ti alloys belong generally to pattern A (Case 1) and B (Case 2), respectively. (6) By applying the equations, f=$(1/{\alpha}_{g1}) sinh^-1({\beta}_{g1}{\dot{s}}$) and $f=(X_{g1}/{\alpha}_{g1})sinh^{-1}({\beta}_{g1}{\dot{s}})+ (X_{g2}/{\alpha}_{g2})\;shih^{-1}({\beta}_{g2}{\dot{s}})$ to the flow data of Fe and Ti alloys, the parametric values of $x_{gj}/{\alpha}_{gj}\;and\;{\beta}_{gs}(j=1\;or\;2)$ were determined, here the subscript g signifies a grain-boundary flow unit. (7) From the values of ($({\beta}_gj)^{-1}$) at different temperatures, the activation enthalpy ${\Delta}H_{gj}^{\neq}$ of deformation due to flow unit gj was determined, ($({\beta}_gj)^{-1}$) being proportional to , the jumping frequency (the rate constant) of flow unit gj. The ${\Delta}H_{gj}\;^{\neq}$ agreed very well with ${\Delta}H_{gj}\;^{\neq}$ (self-diff) of the element j whose diffusion in the sample is a critical step for the deformation as proposed by Kim-Ree's theory (Refer to Tables 3 and 4). (8) The fact, ${\Delta}H_{gj}\;^{\neq}={\Delta}H_{j}\;^{\neq}$ (self-diff), justifies the Kim-Ree theory and their method for determining activation enthalpies for deformation. (9) A linear relation between ${\beta}^{-1}$ and carbon content [C] in hot-rolled steel was observed, i.e., In ${\beta}^{-1}$ = -50.2 [C] - 40.3. This equation explains very well the experimental facts observed with regard to the deformation of hot-rolled steel..

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.9
• /
• pp.1038-1048
• /
• 2019

Recently, various systems have been developed to support ship navigation safety. In order to verify the usefulness of such a system, it is most ideal to try it on a real vessel, but there are many difficulties. As an alternative, usability verification methods applied with modelling and simulation (M&S) techniques are required such as FMSS, which is closest to reality, is very expansive to construct, and there needs the specialized operator. For this reason, this paper proposes a method to verify the navigation safety support system by modeling and simulation techniques based on the Discrete Event System Specification (DEVS) formalism. As a first step, we designed the navigation simulation architecture based on the SES/MB framework, and details on modelling ship core equipment and navigator agents based on the DEVS. Through this, we are able to implement the navigation simulation system for vessels, and evaluate the effectiveness of navigation safety support elements such as collision avoidance, etc. using developed scenarios.