• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.10
• /
• pp.17-22
• /
• 1999

In this paper, the results of the studies about a new proposal where the ONON(oxide-nitride-oxide-nitride) layer instead of the conventional ONO(oxide-nitride-oxide) layer is used as the IPD(inter-poly-dielectrics) layer to improve the data retention problem in the Flash EEPROM cell, have been discussed. For these studies, the stacked-gate Flash EEPROM cell with an about 10nm thick gate oxide and on ONO or ONON IPD layer have been fabricated. The measurement results have shown that the data retention characteristics of the devices with the ONO IPD layer are significantly degraded with an activation energy of 0.78 eV. which is much lower than the minimum value (1.0 eV) required for the Flash EEPROM cell. This is believed to be due to the partial or whole etching of the top oxide of the IPD layer during the cleaning process performed just prior to the dry oxidation process to grow the gate oxide of the peripheral MOSFET devices. Whereas the data retention characteristics of the devices with the ONON IPD layer have been found to be much (more than 50%) improved with an activation energy of 1.10 eV. This must be because the thin nitride layer on the top oxide layer in the ONON IPD layer protected the top oxide layer from being etched during the cleaning process.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.33 no.2
• /
• pp.215-229
• /
• 2023

Objectives: To validate the effectiveness of a real-time chemical exposure monitoring system developed by KOSHA (Korea Occupational Safety and Health Agency), we applied the system to a workplace in the electronics industry for 153 days. Methods: The monitoring system consisted of a PID chemical sensor, a LTE communication equipment, and a web-based platform. To monitor chemical exposure, four sets of sensors were placed in two manufacturing tasks - inspection and jig cleaning - which used TCE as a degreasing agent. We reviewed previous reports of work environment measurements and conducted a new work environment measurement on one day during the period. The PID sensor systems detected the chemical exposure levels in the workplace every second and transmitted it to the platform. Daily average and maximum chemical exposure levels were also recorded. Results: We compared the results from the real-time monitoring system and the work environment measurement by traditional methods. Generally, the data from the real-time monitoring system showed a higher level because the sensors were closer to the chemical source. We found that 28% of jig cleaning task data exceeded the STEL. Peak exposure levels of sensor data were useful for understanding the characteristics of the task's chemical use. Limitations and implications were reviewed for the adoption of the system for preventing poisoning caused by chemical substances. Conclusions: We found that the real-time chemical exposure monitoring system was an efficient tool for preventing occupational diseases caused by chemical exposure, such as acute poisoning. Further research is needed to improve the reliability and applicability of the system. We also believe that forming a social consensus around the system is essential.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.117-127
• /
• 2022

Data anomalies seriously threaten the reliability of the bridge structural health monitoring system and may trigger system misjudgment. To overcome the above problem, an efficient and accurate data anomaly detection method is desiderated. Traditional anomaly detection methods extract various abnormal features as the key indicators to identify data anomalies. Then set thresholds artificially for various features to identify specific anomalies, which is the artificial experience method. However, limited by the poor generalization ability among sensors, this method often leads to high labor costs. Another approach to anomaly detection is a data-driven approach based on machine learning methods. Among these, the bidirectional long-short memory neural network (BiLSTM), as an effective classification method, excels at finding complex relationships in multivariate time series data. However, training unprocessed original signals often leads to low computation efficiency and poor convergence, for lacking appropriate feature selection. Therefore, this article combines the advantages of the two methods by proposing a deep learning method with manual experience statistical features fed into it. Experimental comparative studies illustrate that the BiLSTM model with appropriate feature input has an accuracy rate of over 87-94%. Meanwhile, this paper provides basic principles of data cleaning and discusses the typical features of various anomalies. Furthermore, the optimization strategies of the feature space selection based on artificial experience are also highlighted.

• Water Engineering Research
• /
• v.1 no.1
• /
• pp.39-48
• /
• 2000

This work was completed in partial fulfillment of an on-going research ot descover the effective management of urban nonpoint sources. The current data was obtained from the area of Shingal, Kyunni-do. The investigation was are predominant soures of storm-runoff load and drainage. As a result of the investigation, the road was found to be most seriously contaminated and a significant potential source deteriorating the quality of streams and lakes in the vicinity of the town. Thus, in could be concluded that an effective and systematic cleaning technique must be developed as soon as possible and be frequently applied to the road.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.24 no.5
• /
• pp.772-783
• /
• 2000

The purpose of this study was to investigate product evaluation on consumer's buying behavior of domestic & imported golf wear brands. The questionnaires were sent to 200 consumers who play golf. The 119 data were analyzed by mean, t-test, ANOVA and chi-square. The results of the study were as follows: There were significant differences on consumers' evaluation of apparel quality on fabric and style between groups. Consumers evaluated that the imported golf wears made of more soft, light and unique fabric than domestic, and had a unique and characteristic style. The evaluation of apparel quality according to demographic information has significant difference. Consumers(46-55 ages, business managers and professional) evaluated imported brands were made of soft and light fabric. Consumers(business managers) buying imported brand evaluated dry-cleaning was inconvenient. Consumers who engaged in service industry evaluated domestic brands were easy to coordinate with other items.

• Journal of Environmental Health Sciences
• /
• v.27 no.1
• /
• pp.1-7
• /
• 2001

Ammonia is used in the manufacture of fertilizers, refrigerants, stabilizers and many household cleaning agents. These wide applications resulted in ammonia contamination in water. Ammonia can be removed from water by physical, biological, and chemical methods. Ozonation is effictive in the treatment of water with low concentration of ammonia. This study is undertaken to provide kinetic data for the ozonation of ammonia with or without hydrogen peroxide. The results were as follows; The destruction rate of ammonia increased gradually with the influent hydrogen peroxide concentration up to 0.23 mM and inhibited in the range of 0.23~11.4mM, and the maximum removal rate of ammonia achieved at 0.23mM of hydrogen peroxide, and the overall kinetics was first order. The combination effect of hydrogen and ozone to oxide ammonia in aqueous solution was better than ozone alone. The reacted ammonia was converted completely to nitrate ion.

• Proceedings of the KSME Conference
• /
• 2004.11a
• /
• pp.1629-1633
• /
• 2004

Liquid is commonly introduced as transversal jets in venturi scrubber which is one of the gas cleaning equipments. The jet dynamics such as penetration and breakup is of fundamental importance to the dust-collection efficiency. We have developed a model that can numerically simulate the breakup of the liquid jet in crossflow. This simulation consists of models on liquid column, jet surface breakup, column fracture and secondary droplet breakup. These models have been embedded in the KIVA3-V code. We have calculated such parameters as the jet penetration, jet trajectory, droplet size, velocity field and the volume flux distribution. The results are compared with the experimental data in this paper.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.27 no.2
• /
• pp.200-208
• /
• 2003

The purpose of this study is to propose scientific data and to develop complex\ulcornermultifunctional fibers by applying microcapsules containing rosewood oil on wool fabric. Polyurea was used as wall material of microcapsules. The parameters for adoptable condition are 20min of agitation time, 5000rpm of agitation speed, 4% of emulsifier concentration according to the observation with SEM and particle analyzer. The melting peak of microcapsules was shown to be 280$^{\circ}C$ The amount of released rosewood from microcapsules at 25$^{\circ}C$ was below 1 % and at 40% was below 3%. Therefore, polyurea microcapsules is capable of preserving rosewood oil for long self-life at normal temperature. Antimicrobial activity to dry cleaning and ironing cycle of wool fabric treated with microcapsules were maintained.

• Journal of environmental and Sanitary engineering
• /
• v.17 no.2
• /
• pp.1-10
• /
• 2002

Ammonia is used in the manufacture of fertilizers, refrigerants, stabilizers and many household cleaning agents. The wide applications result in ammonia contamination in water. Ammonia can be removed from water by physical, biological, and chemical methods. Especially ozonation is effective in the treatment of water with low concentration of ammonia. Therefore, this study is undertaken to provide kinetic data for the ozonation of ammonia with bromide. The results were as follows; Ammonia oxidized by ozone with bromide catalysis. The denitrification rate of the ammonia increased proportionally to the concentration of bromide, and the overall reaction order was zero. It was also found that the effect of bromide ion concentration on the denitrification can be expressed by Monod type equation and there was no more effect above a proper bromide ion concentration. The reacted ammonia was converted completely to nitrate ion without bromide, but the denitrification of ammonia by ozone was conducted in the presence of bromide.

• Proceedings of the KAIS Fall Conference
• /
• 2009.12a
• /
• pp.212-213
• /
• 2009

본 연구의 목적은 소방공무원이 현장에서 겪는 충격 스트레스로 발생하는 외상 후 스트레스장애 (posttraumatic stress disorder, PTSD) 분석을 통해 위기상황 스트레스 해소 교육 프로그램 기초 자료를 제시하고자 하는 데 있다. 연구 목적을 달성하기 위하여 독립변인(업무부담감, 스트레스, 스트레스 대응) 3개, 매개변인(현장충격 스트레스) 1개, 종속변인(신체적 증상)으로 구성하여 영향력을 파악하였다. 본 연구 대상자는 전국 970명 소방공무원으로 2007년 3월부터 12월까지 구조화된 설문지를 이용하여 조사하였다. 자료분석은 SPSS 14.0과 구조방정식 모형인 AMOS 7.0 통계패키지를 사용하였으며, 정확한 코딩데이터의 입력확인을 위해 데이터클리닝(data cleaning) 작업을 실시하였다. 가설검증을 위하여 구조방정식 모형분석을 실시한 결과, 업무부담감, 스트레스, 현장충격 스트레스가 낮고, 스트레스 대응이 높을 때 신체적 증상이 낮아지는 것으로 나타났다. 서울소방학교와 공동으로 총 10개 차시로 웹기반 교육을 구성하였으며, 2010년 1월부터 소방공무원 전체를 대상으로 웹기반 교육을 실시하고자 한다.