• 한국환경과학회지
• /
• 제27권11호
• /
• pp.957-966
• /
• 2018

This study intend to induce citizen's voluntary preliminary disaster prevention activity to reduce damage of typhoon that occurs every year. For this purpose, a survey was conducted to develop Typhoon information contents. The number of samples used in the survey was set to 500 people, and citizens living in Jeju, Busan, and Jeonlanam-do were surveyed for areas with high typhoon disasters in order to develop practical and efficient information. The survey consisted of perception about natural disaster, how to get and use weather information, satisfaction with typhoon information and requirements. The general public perceived the typhoon as the first natural disaster. As a result of responding to the method of obtaining and utilizing weather information, the frequency of collecting weather information at the time of issuance of typhoon special report is higher than usual. The purpose of using weather information is clear and the response rate is high for the purpose of disaster prevention. The medium mainly collecting weather information is Internet portal site and mobile phone besides television. The current satisfaction with typhoon weather information is 34.8%, in addition to the accuracy of prediction, it is necessary to improve the information (that is content) provided. Specific responses to the content were investigated not only for single meteorological factors, but also for possible damage and potential countermeasures in the event of a disaster such as a typhoon. As can be seen from the above results, people are requested to provide information that can be used to detect and cope with disasters. The development of new content using easy accessible media will contribute to the reduction of damages caused by the typhoon that will occur in the future, and also to the disaster prevention activity.

• 한국환경과학회지
• /
• 제28권4호
• /
• pp.441-454
• /
• 2019

The present study analyzes the characteristics of 43 typhoons that affected the Korean Peninsula between 2002 and 2015. The analysis was based on 3-second gust measurements, which is the maximum wind speed relevant for typhoon disaster prevention, using a typhoon disaster prevention model. And the distribution and characteristics of the 3-second gusts of four typhoons, RUSA, MAEMI, KOMPASU, and BOLAVEN that caused great damage, were also analyzed. The analysis show that between May and October during which typhoons affected the Korean Peninsula, the month with the highest frequency was August(13 times), followed by July and September with 12 occurrences each. Furthermore, the 3-second gust was strongest at 21.2 m/s in September, followed by 19.6 m/s in August. These results show that the Korean Peninsula was most frequently affected by typhoons in August and September, and the 3-second gusts were also the strongest during these two months. Typhoons MAEMI and KOMPASU showed distribution of strong 3-second gusts in the right area of the typhoon path, whereas typhoons RUSA and BOLAVEN showed strong 3-second gusts over the entire Korean Peninsula. Moreover, 3-second gusts amount of the ratio of 0.7 % in case of RUSA, 0.8 % at MAEMI, 3.3 % at KOMPASU, and 21.8 % at BOLAVEN showed as "very strong", based on the typhoon intensity classification criteria of the Korea Meteorological Administration. Based on the results of this study, a database was built with the frequencies of the monthly typhoons and 3-second gust data for all typhoons that affected the Korean Peninsula, which could be used as the basic data for developing a typhoon disaster prevention system.

• 한국환경과학회지
• /
• 제30권9호
• /
• pp.753-762
• /
• 2021

This study analyzed the characteristics of strong winds accompanying typhoons for a period of 116 years, from 1904 to 2019, when modern weather observations began in Korea. Analysis shows that the average wind speed and high wind rate caused by typhoons were higher over the sea and in the coastal areas than in the inland areas. The average wind speed was higher over the West Sea than over the South Sea, but the rate of strong wind was greater over the South Sea than over the West Sea. The average wind speed decreased by 1980 and recently increased, while the rate of strong winds decreased by 1985 and has subsequently increased. By season, the strong winds in autumn (september and october) were stronger than those in summer (june, july, and august). Strong winds were also more frequent in autumn than in summer. The analysis of the changes in strong winds caused by typhoons since the 1960s shows that the speed of strong winds in august, september, and october has increased more recently than in the past four cycles. In particular, the increase in wind speed was evident in fall (september and october). Analysis of the results suggests that the stronger wind is due to the effects of autumn typhoons, and the increased possibility of strong winds.

• 한국환경과학회지
• /
• 제30권9호
• /
• pp.763-777
• /
• 2021

The spatial characteristics of typhoon-class strong wind during the non-typhoon period were analyzed using, a cluster analysis of the observational data and of special strong wind advisories and, warnings issued by the Korean Meteorological Administration. On the Korean Peninsula, strong winds during non-typhoon periods showed a wide variety of spatial characteristics. In particular, the cluster analysis showed that strong winds could be classified into six clusters on the Korean Peninsula, and that the spatial distribution, occurrence rate of strong winds, and strong wind speed in each cluster were complex and diverse. In addition, our analysis of the frequency of issuance of special strong wind warnings showed a significant difference in the average frequency of strong wind warnings issued in metropolitan cities, with relatively high numbers of warnings issued in Gyeongsangbuk-do and, Jeollanam-do, and low numbers of warning issued inland and in other metropolitan cities. As a result of the changing trend in warnings issued from 2004 to 2019, Ulsan and Busan can be interpreted as having a relatively high number of warnings; the frequency of strong wind warnings issuances and strong wind occurrences in these cities is increasing rapidly. Based on the results of this study, it is necessary to identify areas with similar strong wind characteristics and consider specific regional standards in terms of disaster prevention.

• 한국정보전자통신기술학회논문지
• /
• 제14권4호
• /
• pp.328-337
• /
• 2021

석면, 벤젠과 같이 발암물질 1급인 미세먼지는 각종 질병에 원인이 되고 있다. 초 미세먼지 확산은 코로나 바이러스 확산의 중요한 원인중 하나이다. 본 논문은 2015년부터 2019년까지 서울시 평균 기온, 강수량, 평균 풍속등의 기상 데이터와 SO2, NO2, O3,등의 대기 환경 데이터를 기반으로 미세먼지와 초 미세먼지를 분석하고 예측한다. 계절별과 월별로 미세먼지와 초미세먼지 현황을 파악·분석하며 미세먼지를 예측하기 위해 기계학습 모델 중 선형회귀, SVM, 앙상블 모델을 이용하여 비교 분석하였다. 또한 미세먼지와 초 미세먼지 발생에 영향을 미치는 중요한 피쳐(속성)를 파악한다. 본 논문이 파악한 결과 3월에 가장 (초)미세먼지가 높았고, 8월에서 9월까지 (초)미세먼지가 가장 낮았다. 기상 데이터일 경우 (초)미세먼지에 가장 영향을 미치는 데이터가 평균 기온이며, 기상 데이터와 대기 환경 데이터일 경우 NO2가 (초)미세먼지 발생에 가장 크게 작용하였다.

• 한국환경과학회지
• /
• 제31권1호
• /
• pp.87-97
• /
• 2022

In this study, type analysis was conducted along with the advancement of basic data to calculate the maximum damage caused by strong winds during the typhoon period. The result of the damage by region showed that in 2012, the difference in damage was clearly distinguished as the region was classified in detail. In addition, the result of the annual damage in 2011 was strong on the west coast, and in 2016, the damage to the southeast coast was significant. In 2012, the 3-second gust was relatively stronger on the west and southeast coasts than in 2011, and the winds blew stronger along the southeast coast in 2016. Monthly damage data showed that the damage to the west coast was high in August, and the damage to the southeast coast was high in October from 2002 to 2019. The 3-second gust showed the result of wide expansion throughout the southern coast of the Korean Peninsula in October. As a result, the damage differs for type bacause the intensities and paths of typhoons vary depending on their characteristics, the 3-second gust blows differently by region based on regional characteristics, and the sale price is considered in metropolitan cities.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권5호
• /
• pp.1431-1445
• /
• 2022

We construct a deep neural network model named ECGResNet. This model can diagnosis diseases based on 12-lead ECG data of eight common cardiovascular diseases with a high accuracy. We chose the 16 Blocks of ResNet50 as the main body of the model and added the Squeeze-and-Excitation module to learn the data information between channels adaptively. We modified the first convolutional layer of ResNet50 which has a convolutional kernel of 7 to a superposition of convolutional kernels of 8 and 16 as our feature extraction method. This way allows the model to focus on the overall trend of the ECG signal while also noticing subtle changes. The model further improves the accuracy of cardiovascular and cerebrovascular disease classification by using a fully connected layer that integrates factors such as gender and age. The ECGResNet model adds Dropout layers to both the residual block and SE module of ResNet50, further avoiding the phenomenon of model overfitting. The model was eventually trained using a five-fold cross-validation and Flooding training method, with an accuracy of 95% on the test set and an F1-score of 0.841.We design a new deep neural network, innovate a multi-scale feature extraction method, and apply the SE module to extract features of ECG data.

• 한국대기환경학회지
• /
• 제28권2호
• /
• pp.211-221
• /
• 2012

The Ministry of Environment will enforce air quality standards for $PM_{2.5}$ in 2015 because it affects human health as well as climate change and brings about other adverse effects. Until recently, even though a number of researches have reported $PM_{2.5}$ emissions according to sources, they have not precisely considered the emission factors correspondent to each source for emission estimation. For the sake of establishing $PM_{2.5}$ emission inventories, this study was undertaken using activity data of each source taken from CAPSS (Clean Air Policy Support System) multiplied by each emission factor of U.S. EPA Speciate or EEA CORINAIR. The 2008's total annual $PM_{2.5}$ emission (73.971 ton/yr) can be apportioned into five sources, such as fuel combustion facilities (62.2%), mobiles (33.8%), production processes (3.2%), fires (0.4%), and waste treatments (0.3%). The results show that fuel combustion facilities and mobiles are the predominant sources of $PM_{2.5}$, and they should be taken into great account in establishing $PM_{2.5}$ standards. In addition, it is necessary and urgent to develop effective measures for reduction of $PM_{2.5}$ emissions from those two main sources as well.

• 한국대기환경학회:학술대회논문집
• /
• 한국대기환경학회 1997년도 한국대기보전학회 추계학술대회 요지집
• /
• pp.98-99
• /
• 1997

• 한국대기환경학회:학술대회논문집
• /
• 한국대기환경학회 1996년도 한국대기보전학회 추계학술대회 요지집
• /
• pp.3-4
• /
• 1996