• 한국대기환경학회지
• /
• 제16권5호
• /
• pp.445-451
• /
• 2000

To estimate dry deposition flux of 12 elements in aerosols, aerosol particles were sampled by a low-pressure impactor(LPI) and a dust jar. The concentrations of 12 elements in aerosol particle and dry deposition were analyzed by a PIXE analysis using as a 2.0 MeV-proton beam. The mean dry deposition velocities of 12 elements were estimated by ranges of 0.74∼2.62 cm/sec. The results showed that the highest value was 3.26 cm/sec for Ca and the lowest value 0.74 cm/sec for Fe. The dry deposition flux for elements was calculated as a function of particle size by 1-step method and 12-step method. In this work, dry deposition velocities were computed with the two existing models; the coarse-particle fraction(4∼30 mm diameter) using the dry deposition velocity model of the Noll and Fang(1998) and the fine-particle fraction (0.05∼4mm diameter) using the Shemel and Hodgson(1980) model. The ratios of the mean calculated/measured fluxes were 3.59 for 1-step method and 0.60 for 12-step method respectively.

• Structural Monitoring and Maintenance
• /
• 제5권1호
• /
• pp.51-65
• /
• 2018

A number of sensing techniques have been implemented for detecting defects in civil infrastructures instead of onsite human inspections in structural health monitoring. However, the issue of faults in sensors has not received much attention. This issue may lead to incorrect interpretation of data and false alarms. To overcome these challenges, this article presents a deep learning-based method with a new architecture of Stateful Long Short Term Memory Neural Networks (S-LSTM NN) for detecting sensor fault without going into details of the fault features. As LSTMs are capable of learning data features automatically, and the proposed method works without an accurate mathematical model. The detection of four types of sensor faults are studied in this paper. Non-stationary acceleration responses of a three-span continuous bridge when under operational conditions are studied. A deep network model is applied to the measured bridge data with estimation to detect the sensor fault. Another set of sensor output data is used to supervise the network parameters and backpropagation algorithm to fine tune the parameters to establish a deep self-coding network model. The response residuals between the true value and the predicted value of the deep S-LSTM network was statistically analyzed to determine the fault threshold of sensor. Experimental study with a cable-stayed bridge further indicated that the proposed method is robust in the detection of the sensor fault.

• 한국가축번식학회지
• /
• 제12권1호
• /
• pp.24-30
• /
• 1988

These experiments were carried out to develop new techniques for In Vitro separatin of X-and Y-bearing spermatozoa. The bull semen was applied to the various Gel-Columns filled with swellen Sephadex G-50 Fine and then elutriated wtih Locke solution (elutriation rate; 1ml/3-4min., 1ml/1-2min.). Elutriated solution was fractionated into 1ml by automatic Fraction Collector and spermatozoa included in each fraction were subjected to the estimation of viability and recovery rate, and to B-body test. The results obtained in these experiments were summarized as follows: 1. When the column size and the elutriation rate were adjusted to 15$\times$1.6cm and 1ml/3-4min., respectively, the highest sperm concentration was obtained from the 8th to the 12th fraction. 2. As a trend, the viability of spermatozoa was improved by chromatography, and the degree of improvement ranged 5 to 10 percentage. 3. The average recovery rate of spermatozoa applied to column was 73.2 percentage and ranged 52.6 to 81.3 percentage. 4. The lowest rate of B-body bearing spermatozoa following chromatography was obtained when the column size and the elutriation rate were adjusted to 15$\times$0.8cm and 1ml/1-2min., respectively.

• Journal of Astronomy and Space Sciences
• /
• 제33권1호
• /
• pp.37-44
• /
• 2016

This paper presents an algorithm for Real-Time Orbit Determination (RTOD) of navigation satellites for the Korean Regional Navigation Satellite System (KRNSS), when the navigation satellites generate ephemeris by themselves in abnormal situations. The KRNSS is an independent Regional Navigation Satellite System (RNSS) that is currently within the basic/preliminary research phase, which is intended to provide a satellite navigation service for South Korea and neighboring countries. Its candidate constellation comprises three geostationary and four elliptical inclined geosynchronous orbit satellites. Relative distance ranging between the KRNSS satellites based on Inter-Satellite Ranging (ISR) is adopted as the observation model. The extended Kalman filter is used for real-time estimation, which includes fine-tuning the covariance, measurement noise, and process noise matrices. Simulation results show that ISR precision of 0.3-0.7 m, ranging capability of 65,000 km, and observation intervals of less than 20 min are required to accomplish RTOD accuracy to within 1 m. Furthermore, close correlation is confirmed between the dilution of precision and RTOD accuracy.

• 한국안전학회지
• /
• 제25권1호
• /
• pp.32-37
• /
• 2010

When back-out & firing Process applies heat, hume is piled up in exhaust duct by organic compound and it have high dangerousness. There by, the process is happening a lot of damage that is exhaust duct fire. However we do not have certain fire dangerousness estimation and digestion countermeasure. So we need preventive measure. Back-out & firing is a process which has fine structure, electrical and mechanical characteristics, such as firing kiln and back-out kiln which has pipe line and box type. The box oven is made of heating coil, fan motor and control panel. Back-out & firing process has air circulation institution of quick ventilation type. When we operate this process for long time, fire can break out easily. Duct is made by zinc shredder. If fire breaks out in duct inside, fire by deposit fume can be dispersed easily. Accordinglym, This project estimate danger for back-out & firing process exhaust duct through real fire test. And there is purpose of study to establish preventive measure.

• 전자공학회논문지CI
• /
• 제41권6호
• /
• pp.47-53
• /
• 2004

본 논문에서는 외향 기반 접근법을 기반으로 한 로봇의 위치 추정 알고리즘을 제안한다. 우선, 제안한 알고리즘은 주성분 분석(PCA: Principal Component Analysis)을 이용하여 취득한 영상들을 eigenspace로 투영시킴으로써 영상을 압축한다. 추출된 주성분은 eigenspace에서의 연속 외향 함수(continuous appearance function)로 나타낼 수 있다. 신경 회로망은 로봇의 위치추정을 위해 새로운 영상이 주어지면 이것을 eigenspace로 투영 시킨 후 연속 외향 함수를 통해 로봇의 현재 위치를 추정한다. 최종적으로는, 영상안의 데이터에 칼만 필터를 적용함으로써 로봇의 정확한 위치를 추정할 수 있다. 제안한 알고리즘을 실제 이동 로봇에 탑재하여 적용시킨 결과 로봇의 위치를 정확히 추정할 수 있음을 확인 할 수 있었다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2005년도 춘계 학술발표회 논문집
• /
• pp.1159-1166
• /
• 2005

In the present study, a series of laboratory tests with sands of different silt contents, are conducted and methods to assess non-linear behaviors based on in-situ test results are proposed. Modified hyperbolic stress-strain model is used to analyze non-linearity of silty sands in terms of non-linear degradation parameters f and g as a function of silt contents and relative density $D_R$. Stress-strain relationship results were obtained from a series of triaxial tests on sands containing different amounts of silt. Initial shear modulus which was applied to normalize modulus degradation of silty sands were determined based on the resonant column test results. From the laboratory test results, it was observed that, as the relative density increases, values of f decrease and those of g increase. Cone resistance $q_c$ for silty soil condition used in the triaxial tests were estimated based on the cavity expansion analysis. A suggestion to make an estimation of degradation parameters f and g as a function of fine contents is addressed in terms of cone resistance $q_c$ .

• BMB Reports
• /
• 제38권5호
• /
• pp.555-562
• /
• 2005

Single nucleotide polymorphisms (SNPs) are becoming the most common type of markers used in genetic analysis. In the present report a SNP has been chosen to test the applicability of Real Time PCR to discriminate and quantify SNPs alleles on DNA pools. Amplification Refractory Mutation System (ARMS) and Mismatch Amplification Mutation Assay (MAMA) has been applied. Each assay has been pre-validated testing specificity and performances (linearity, PCR efficiency, interference limit, limit of detection, limit of quantification, precision and accuracy). Both the approaches achieve a precise and accurate estimation of the allele frequencies on pooled DNA samples in the range from 5% to 95% and don't require standard curves or calibrators. The lowest measurement that could be significantly distinguished from the background noise has been determined around the 1% for both the approaches, allowing to extend the range of quantifications from 1% to 99%. Furthermore applicability of Real Time PCR assays for general diagnostic purposes is discussed.

• 한국대기환경학회지
• /
• 제11권3호
• /
• pp.263-272
• /
• 1995

Characteristics of anbient at Korean, Cheju Island have been studied during the intensive field study period on March 11 .sim. 17, 1994 in collaboration with other research organizations from Korea and abroad. The particle size distribution was measured using an Electrical Aerosol Analyzer(EAA) and an Optical particle Counter(OPC). Fine particles(PM1 and PM3) have been collected by filter pack samplers and their ionic compositions have been analyzed. sampling errors inherent to the filter pack sampling method are discussed and the method to analyze those errors are presented. The rine mass concentrations of this study show very similar mass concentrations when Seoul is clear. This is somewhat surprising result, because the most of researchers believe that Kosan is one of the cleanest area in Korea. Bimodal volume size distributions with peak values around 0.1 .sim. 0.2.mu.m and 3.mu.m in particle dimeter were observed for most of the measurement period, particle mass loadings and ionic composition data show a large fraction of particles are from non-sea salt origins. Estimation of water content and acidity of particles based on measurement by a gas/particle equilibrium model, SCAPE, reveals that the pH values of particles are comparable to or lower than those estimated based on measurements in Los Angeles, U.S.A. during the SCAQS study. These findings with the meteorological conditions during the study period suggest that the particles collected during the period have originated from outside Cheju Island.

• 한국대기환경학회지
• /
• 제11권3호
• /
• pp.279-290
• /
• 1995

Recently in Korea, due to the significant drop of lead and bromine levels as a marker of autoemission source in the urban areas, the conventional application of receptor methods has many difficulties to properly apportion mass contribution of some sources. It is then needed to urgently develop alternative source profiles and identify new emission markers. Thus, the study has extensively examined the results obtained from using PAHs and elemental data for receptor modeling and has provided an opportunity to identify alternative source compositions and to determine a proper number of the ambient emission sources in Seoul area. The purpose of the study is to identify the sources of PM-10 and to estimate their mass contributions in Seoul area. Thus, a receptor model, target transformation factor analysis(TTFA) has been massively applied. The TTFA offers the possibility of determining the number of sources and their mass contributions. The input data used in this study are composed of two separate sets: fine (d$_{p}$ < 2.5.mu.m) and coarse (2.5.mu.m < d$_{p}$ < 10.mu.m) mode aerosol samples. Each sample was simultaneously collected by a PM-10 dichotomous sampler during the daytime(8 AM to 8 PM) and the nighttime(8 PM to 8 AM) from February to October 1993 on the Sungdong-Gu, Seoul. All the samples were analyzed to determine the levels of 10 inorganic elements by an XRF system as well as 14 PAHs by a HPLC. However, only 8 inorganic elements and 7 PAHs were used for the various statistical analysis.sis.