• Journal of Intelligence and Information Systems
• /
• v.4 no.1
• /
• pp.79-87
• /
• 1998

정확한 태풍진로 예측은 동아시아 최대의 자연재해인 태풍의 피해를 최소화하는데 필수적이다. 기상역학에 기초를 둔 수치모델과 회귀분석등의 통계적 접근법이 사용되어왔다. 본 논문에서는 비선형 신경망모델인 다층퍼셉트론을 제안한다. 즉, 태풍진로예측을 이동경로, 속도, 기압 등의 변수로 이루어진 시계열의 예측으로 본다. 1945년부터 1989년까지 한반도에 접근한 태풍 데이터를 이용하여 제안된 신경망을 학습한 후, 94, 95년도에 접근한 태풍의 진로를 예측하였다. 신경망의 예측성능은 수치모델의 성능보다 조금 우수하거나 비슷하였다. 신경망의 성능은 충분히 더 향상될 수 있는 여지가 있다. 또한, 고가의 슈퍼컴퓨터로 여러 시간 계산을 해야하는 수치모델에 비하여 PC상에서 수초만에 계산을 할 수 있는 신경망 모델은 비용 면에서도 장점이 있다.

• Atmosphere
• /
• v.24 no.4
• /
• pp.565-571
• /
• 2014

A long-range prediction system of tropical cyclone (TC) activity over the western North Pacific (WNP) has been operated in the National Typhoon Center of the Korea Meteorological Administration since 2012. The model forecasts the spatial distribution of TC tracks averaged over the period June~October. In this study, we separately developed TC prediction models for summer (June~August) and autumn (September~November) period based on the current operating system. To perform the three-month WNP TC activity prediction procedure readily, we modified the shell script calling in environmental variables automatically. The user can apply the model by changing these environmental variables of namelist parameter in consideration of their objective. The validations for the two seasons demonstrate the great performance of predictions showing high pattern correlations between hindcast and observed TC activity. In addition, we developed a post-processing script for deducing TC activity in the Korea emergency zone from final forecasting map and its skill is discussed.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.369-369
• /
• 2021

기후변화로 인하여 태풍의 발생 횟수가 증가하고, 태풍 크기가 점차 대형화되며, 강도 또한 강해지는 추세에 있다. 우리나라도 매년 7월에서 10월 사이에 태풍으로 인한 피해가 발생하나, 태풍피해의 빈도나 피해 규모는 일정하지 않으며, 이는 태풍 진로와 태풍 크기 및 강도와 관계있다. 태풍에 대한 분류는 태풍의 크기와 중심부로 불어오는 풍속에 따라 강도로 구분하며, 태풍의 크기에 따른 분류는 풍속 15m/sec 이상 되는 영역의 반경에 따라 소형(300km 미만), 중형(500km미만), 대형(800km미만), 초대형(800km이상) 등 4계급 구간으로 구분하고, 태풍의 강도는 17m/s~25m/s 범위내의 태풍은 강도를 정하지 않으며, 중(25m/s~33m/s), 강(33m/s~44m/s), 매우강(44m/s~54m/s), 초강력(54m/s 이상)으로 구분한다. 최근 10년간 자연재해 중 태풍으로 인한 피해는 1조 6825억원으로 우리나라 자연재해 총피해액인 3조 6280억의 46%를 차지하며, 원인별로 가장 큰 피해를 야기하며, 또 태풍 루사, 매미는 단일 재해로는 최대규모로 알려져 있다. 태풍으로 인한 재해는 호우, 강풍, 풍랑으로 인한 피해가 동시에 발생하기 때문이며, 재해에 대한 대비 활동도 복합적으로 이루어져야 한다. 재해예방 측면에서 재해가 우려되는 기상 상황(호우, 강풍, 태풍 등)이 예측되고, 예측된 기상상황 하에서 피해 정도를 추정할 수 있다면 재해 예방을 위하여 적절한 대비를 취할 수 있을 것이다. 태풍은 적도부근 태평양에서 발생하여 이동하는데, 이동경로와 태풍강도는 기상 상황에 따라 변동이 심하므로, 태풍으로 인한 재해를 예측하고 예방하기 위한 대비에도 어려움이 있다. 또 태풍에 대한 기상특보는 태풍의 진로, 크기, 강도를 중심으로 강우량과 최대풍속이 예보되는데, 이것만으로 피해정도를 예측하는데 어려움이 있다. 본 연구에서 우리나라에 직접적인 영향을 미친 태풍을 대상으로, 태풍시 발생한 호우와 풍속이 태풍으로 인한 피해 규모와 관련이 있는 지 여부를 평가하고, 이들 관계를 밝히고자 한다.

• Proceedings of KOSOMES biannual meeting
• /
• 2017.11a
• /
• pp.242-242
• /
• 2017

• Journal of the Korea Institute of Building Construction
• /
• v.16 no.5
• /
• pp.437-445
• /
• 2016

Since typhoon is a critical meteorological disaster, some advanced countries have developed typhoon damage prediction models. However, although South Korea is vulnerable to typhoons, there is still shortage of study in typhoon damage prediction model reflecting the vulnerability of domestic building and features of disaster. Moreover, many studies have been only focused on the characteristics and typhoon and regional characteristics without various influencing factors. Therefore, the objective of this study is to analyze typhoon damage by path and develop to prediction model for building damage ratio by using multiple regression analysis. This study classifies the building damages by typhoon paths to identify influencing factors then the correlation analysis is conducted between building damage ratio and their factors. In addition, a multiple regression analysis is applied to develop a typhoon damage prediction model. Four categories; typhoon information, geography, construction environment, and socio-economy, are used as the independent variables. The results of this study will be used as fundamental material for the typhoon damage prediction model development of South Korea.

• Journal of the Korean earth science society
• /
• v.30 no.3
• /
• pp.294-304
• /
• 2009

A statistical prediction model for the typhoon intensity and track in the Northwestern Pacific area was developed based on the artificial neural network scheme. Specifically, this model is focused on the 5-day prediction after tropical cyclone genesis, and used the CLIPPER parameters (genesis location, intensity, and date), dynamic parameters (vertical wind shear between 200 and 850hPa, upper-level divergence, and lower-level relative vorticity), and thermal parameters (upper-level equivalent potential temperature, ENSO, 200-hPa air temperature, mid-level relative humidity). Based on the characteristics of predictors, a total of seven artificial neural network models were developed. The best one was the case that combined the CLIPPER parameters and thermal parameters. This case showed higher predictability during the summer season than the winter season, and the forecast error also depended on the location: The intensity error rate increases when the genesis location moves to Southeastern area and the track error increases when it moves to Northwestern area. Comparing the predictability with the multiple linear regression model, the artificial neural network model showed better performance.

• Proceedings of the KIEE Conference
• /
• 2001.11c
• /
• pp.446-449
• /
• 2001

본 논문에서는 태풍의 진로와 세기를 ES_BLRNN을 이용해 예측하였다. 기존의 방법인 수치모델이나. CLIPER모델을 사용함에 있어서, 통계적 방법인 CLIPER모델은 예측성능면에서 수치모델보다 그 성능이 떨어지고, 반면에 수치모델의 성능은 CLIPER 모델에 비해 우수하나 슈퍼컴퓨터(Cray-2S, FUSITSU)를 이용하여야만 예보가 가능한 제약점을 가지고 있다. 또한 수치모델을 슈퍼컴퓨터로 계산할 경우 약 30분 정도가 소요되는 점을 감안할 때, ES_BLRNN은 이들의 단점을 보안할 수 있는 하나의 방편이라 생각된다. 게다가 ES_BLRNN의 경우 개인용 컴퓨터로도 충분히 사용 가능할 만큼 비용이 저렴하고, 681개의 태풍을 학습할 때 결리는 시간은 약 5분 정도이며, 146개의 태풍을 예측하는데 걸리는 시간은 약 3초 정도(Pentium MMX 200 Processor, RAM 64m, OS: RedHat LINUX 5.2. language ; ANSI-C)로써, 슈퍼컴퓨터나 CLIPER모델에 비해 훨씬 빠르게 결과를 볼 수 있다.

• Atmosphere
• /
• v.24 no.1
• /
• pp.123-130
• /
• 2014

The automated prediction system for seasonal tropical cyclone (TC) activity is established at the National Typhoon Center of the Korea Meteorological Administration (KMA) to provide effective operation and control of the system for user who lacks knowledge of the system. For automation of the system, two procedures which include subjective decisions by user are performed in advance, and their output data are provided as input data. To provide the capability to understand the operational processes for operational user, the input and output data are summarized with each process, and the directory structure is reconstructed following KMA's standard. We introduce a user interface using namelist input parameters to effectively control operational conditions which is fixed or should be manually set in the previous version of the prediction system. To operationally use early prediction which become available through the automation, its performances are evaluated according to initial condition dates. As a result, high correlations between the observed and predicted TC counts are kept for all track clusters even though advancing the initial condition date from May to January.

• Journal of the Korean earth science society
• /
• v.35 no.3
• /
• pp.168-179
• /
• 2014

The purpose of this study focuses on the prediction time and location of turning-point of typhoon tracks using the water vapor images of Communication, Ocean and Meteorological Satellite (COMS) which has a very short observation interval. It targets a more accurate prediction of turning-point of typhoon tracks through the relationship between dry slot and northern/southern oscillations of jet stream. Jet stream moves by the position of jet streak and the ${\upsilon}$-component velocity of geostrophic wind. If the ${\upsilon}$-component of geostrophic wind gets stronger toward south, jet stream develops into a circular jet. In that condition, dry slot in satellite water vapor imagery extends toward south, and typhoon track turns as the distance of curved moisture band (CMB) gets narrowed down. If the interval of CMB is below $15^{\circ}$ of latitude, the typhoon track is turning toward north or northeast within 24 hours. As a result, typhoon track showed that when dry slot position was located less than $32^{\circ}N$, typhoon turned its track at $20-23^{\circ}N$ ($1^{th}$ Kong-Rey 2007 and $17^{th}$ Jelawt at 2012), and when in $35^{\circ}N$ above, it turned at $27^{\circ}N$ ($4^{th}$ Man-yi 2007).

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.19 no.5
• /
• pp.492-501
• /
• 2007

Coastal disasters have become one of the most important issues in every coastal country. In Korea, coastal disasters such as storm surge, sea level rise and extreme weather have placed many coastal regions in danger of being exposed or damaged during subsequent storms and gradual shoreline retreat. A storm surge is an onshore gush of water associated with a tow pressure weather system, typically in typhoon season. However, it is very difficult to predict storm surge height and inundation due to the irregularity of the course and intensity of a typhoon. To provide a new scheme of typhoon damage prediction model, the scenario which changes the central pressure, the maximum wind radius, the track and the proceeding speed by corresponding previous typhoon database, was composed. The virtual typhoon scenario database was constructed with individual scenario simulation and evaluation, in which it extracted the result from the scenario database of information of the hereafter typhoon and information due to climate change. This virtual typhoon scenario database will apply damage prediction information about a typhoon. This study performed construction and analysis of the simulation system with the storm surge/coastal inundation model at Masan coastal areas, and applied method for predicting using the scenario of the storm surge.