• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2017.11a
• /
• pp.366-368
• /
• 2017

본 연구는 국내 산사태 발생 데이터를 기반으로 시뮬레이션 모델을 머신러닝 기법을 통해 학습시켜 산사태의 토석류 흐름을 구현하는 알고리즘에 대한 연구이다. 전통적인 프로그래밍을 통한 산사태 시뮬레이션 모델 개발을 해당 시스템에 더 많은 고도의 물리학 법칙을 통합 적용시켜 토석류의 흐름을 공학적으로 재현해내는데 중점을 두고 개발이 진행되지만, 본 연구에서 다루는 머신러닝 기법을 통한 산사태 시뮬레이션 모델 개발의 경우 시스템에 입력되는 데이터를 기반으로한 학습을 통하여 토석류 흐름에 영향을 미치는 변수와 파라메터를 산출하고 정의는데 중점을 두고 개발이 진행된다. 본 연구에서 산사태 시뮬레이션 모델 개발에 활용하는 머신러닝 알고리즘은 강화학습 알고리즘으로 기존 산사태 발생 지점을 기반으로 에이전트를 설정해 시간에 따라 시뮬레이션의 각 스텝에서 토석류의 흐름 즉 액션을 환경에 따른 가중치를 기준으로 산정하게 된다. 여기서 환경에 따른 가중치는 시뮬레이션 모델에 정의된 메서드에 따라 산정된다. 시간이 목표값에 도달하여 결과가 출력되면 출력된 결과와 해당 산사태 발생 지점의 실제 산사태 피해 지역 데이터 즉 시뮬레이션 결과 이상치와의 비교를 통하여 시뮬레이션을 평가하게 된다. 이러한 평가는 시뮬레이션 데이터와 실제 데이터간의 유사도 비교를 통해 손실률을 도출하게 되고 이러한 손실률을 경사하강법등의 최적화 알고리즘을 통해 최소화 하여 입력된 데이터를 기반으로한 최적의 토석류 흐름 구현 알고리즘을 도출한다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.11 no.2
• /
• pp.53-58
• /
• 1993

Landslides, failure of slope stability by natural or artificial factors, occur loss of life and properties. Recently, landslides hazard area predict statistical methods and field measurements, but there are so many difficulties to find the occurrence system because of its complexity. To predict the landslide harvard region, model area is the Seodong in Pusan where occurred landslides. Database of ground height made the each topography in map scale of 1 : 25,000, 1 : 10,000, 1 : 5,000 and 1 : 1,200. Authors knew to landslide hazard area by the weight of ground height data and slope angle data. Finally, aerial photo analysis is possible find landslide hazard area.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.4
• /
• pp.107-119
• /
• 2003

Serious damages occurred in Yeongdong area of Gangwondo by typoon Rusa in August, 2002. Therefore, this study intends to investigate and analyse the characteristics of landslide having occurred in Gangneung area. 11 sites including many features of landslide were selected and scales, soil properties and a state of vegatation of slopes were investigated. It can be concluded that rainfall intensity, the velocity of the wind, geomorphology, vegatation and forest fire etc. are considered as main factors affecting occurrence of landslides in this area.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.4
• /
• pp.81-90
• /
• 2005

Based on the landslide hazard scoring system of Korea Forest Research Institute, a GIS model for predicting landslide hazards was developed. The risk of landslide hazards was analyzed as the function of 7 environmental site factors for the terrain, vegetation, and geological characteristics of the corresponding forest stand sites. Among the environmental factors, slope distance, relative height and shapes of slopes were interpreted using the forestland slope interpretation module developed by Chung et al. (2002). The program consists of three modules for managing spatial data, analyzing landslide hazard and report-writing, A performance test of the model showed that 72% of the total landslides in Youngin-Ansung landslides area took place in the highly vulnerable zones of grade 1 or 2 of the landslide hazard scoring map.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.2
• /
• pp.54-66
• /
• 2006

Korea Forest Service made the landslide hazard map for all mountainous districts over the country in May 2005. In this study, we selected landslide areas occurred in Jeonbuk from 02 August 2005 to 03 August 2005 as the study area. We extracted landslide areas using images taken by PKNU 3 System, which was developed by PE&RS Laboratory in Dept. of Satellite Information Sciences, Pukyong National University and verified the accuracy of landslide hazard map by overlaying landslide hazard areas extracted by PKNU 3 images. And we analyzed characteristics of an altitude, a gradient, an inclined direction, a flow length, a flow accumulation for landslide areas using mountainous terrain analysis and Stream Network analysis of ArvView 3.3. As a result of this study, it is necessary to adjust the unitage(%) by the class and to modify and improve the score table for prediction of landslide-susceptible area forming the foundation of making the landslide hazard maps.

• Journal of Korean Society for Geospatial Information Science
• /
• v.15 no.4
• /
• pp.11-20
• /
• 2007

Most natural disasters in Korea are caused by meteorological natural phenomena, which include storms, heavy rains, heavy snow, hail, tidal waves, and earthquakes. Rainfall is the most frequent cause of disasters and accounts for about 80% of all disasters. Particularly in recent years, Korea has seen annual occurrences of natural disasters associated with landslides (slope and retaining wall collapse and burying) due to meteorological causes from the increasing intensity of heavy rains including local heavy rainfalls. In Korea, it is critical to analyze the characteristics of landslides according to rainfall characteristics and to take necessary and proper measures for them. This study assessed the possibility of landslides in the Gangwon region with a geographic information system by taking into account the inducer factors of landslides and the maximum continuous rainfall of each area. It also analyzed areas susceptible to landslides and checked the distribution of landslide-prone areas by considering the rainfall characteristics of those areas.

• Land and Housing Review
• /
• v.2 no.2
• /
• pp.157-162
• /
• 2011

This paper presents the landslide characteristics occurred in Hoengseong, Gangwondo and around July 16 in 2006, caused by heavy rainfall and antecedent precipitation by two typhoons of Ewiniar and Bilis. The main causes of landslides were antecedent precipitation between July 12 to 13, resulting in weakening grounds by increasing the degree of saturation previously, and the additional heavy rainfall between July 15 to 16. Most of landslides at natural slopes were transitional failures occurred along the boundary between residual weathered soil in shallow depth and hard mother rock. From the results of conclusive analyses for 100 sites in Hoengseong region where landslides occurred, the slope length of landslide and slope width were less than 100m with 87% of frequency and 30m with 74% of frequency, respectively. The average value of slope angles was $24^{\circ}$.

• Economic and Environmental Geology
• /
• v.44 no.4
• /
• pp.289-302
• /
• 2011

Since the frequency of the earthquake occurrence in Korean peninsular is continuously increasing, the possibility that massive landslides are triggered by earthquake is also growing in Korea. Previously, the landslide is known to be induced by large magnitude earthquake, whose magnitude is larger than 6.0. However, the landslide can be induced by only small magnitude earthquake, especially in the fully saturated soil. Therefore, the susceptibility of landslide caused by small magnitude earthquake in fully saturated soil is analyzed in this study. For that, the topographical and geological characteristics of the site were obtained and managed by GIS software. In the procedure of the study, slope angle, cohesion, friction angle, unit weight of soil were obtained and constructed as a spatial database layer. Combining these data sets in a dynamic model based on Newmark's displacement analysis, the landslide displacements were estimated in each grid cell. In order to check out the possibility of the earthquake induced landslides, the level of the groundwater table is varied from dry to 80% saturated soil. In addition, in order to analyze the effect of the magnitude of earthquake and distance to epicenter, four different earthquakes epicenters were considered in the study area.

• The Journal of Engineering Geology
• /
• v.15 no.4 s.42
• /
• pp.435-445
• /
• 2005

Landslides induced by heavy rainfall from typhoon 'Rusa' in 2002 and typhoon 'Meami' in 2003 were investigated at Samcheok area, and the relationship between landslides and rainfall on that area was analyzed. The average annual rainfall at Samcheok area is generally $1,200mm\~1,300mm$. However, the average annual rainfall at samcheok for 2003 and ton was increased more than 2,000mm because of typhoon 'Rusa' and typhoon 'Meami'. The number of landslides and the landslides area are largely occurred in a area of the relatively high maximum hourly rainfall and 2days cumulative rainfall. Therefore, it confirmed that landslides are directly depended on the hourly rainfall and the cumulative rainfall. The landslides at Samcheok area induced by heavy rainfall due to typhoon are more influenced by the maximum hourly rainfall at the landslide occurrence day. In order to predict a rational landslide size, a new method included the maximum hourly rainfall and the landslide area in a traditional way was proposed. As the result of applying the new proposed method, the landslide size at Samcheok area is involved in the large scale landslide.

• Journal of Korean Society for Geospatial Information Science
• /
• v.14 no.1 s.35
• /
• pp.3-12
• /
• 2006

Kangwondo area is mountainous and landslide happens easily during the rainy period in summer time. Especially, when there is torrential downpour caused by the unusual weather change, there will be greater possibility to see landslide. It is very difficult to analyze and study a natural phenomenon like the landslide because there are so many factors behind it. And the way to conduct the analysis is also very complicated. However, if GIS is used, we can classify and analyze data efficiently by modeling the real phenomenon with a computer. Based upon the analysis on the causes of landslide in the areas where it occurred in the past, therefore, this study shows several factors leading to landslide and contains the GIS database categorized by grade and stored in the computer. In order to analyze the influence of every factor causing landslide, we calculated the rates of weight by AHP and evaluated landslide vulnerability in the study area by using GIS. As a result of such analysis, we found that the forest factor has most potential influences among other factors in landslide.