http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Deying Li,Fasheng Miao,Yuanhua Xie,Chin Leo 대한토목학회 2019 KSCE JOURNAL OF CIVIL ENGINEERING Vol.23 No.12
Landslide hazard prediction for the Three Gorges area is necessary for mitigating geohazards, especially under extreme rainfall conditions. A method for calculating rainfall in the extreme rainfall return period was first proposed. Then, geological conditions of the Baishuihe landslide on the reservoir shore were modelled, along with soil parameters. Four geological profiles were chosen, and the phreatic line of the landslide was simulated in the SEEP/W programme. The profiles were then divided into slices, and, the long-term stability and failure probability of each slice was calculated using the uncertainty of the soil parameters and the Monte Carlo method. An Empirical Bayesian Kriging (EBK) method in ArcGIS was used to obtain a hazard distribution map for the landslide’s active area and the deeper landslide. The standardized extreme rainfall for different return periods were then used to predict the hazard of the active area and determine the relationship between the unstable area within the active area and the rainfall return period. The stability of the Baishuihe landslide shows a periodic trend and a strong relationship with the reservoir water level and the rainfall distribution, while the stability of the deeper landslide is less affected. With an increase in the rainfall return period, the unstable area of the active area expands. The ratio of the unstable area in the active area and the rainfall return period show a logarithmic correlation. This paper solves the standardization problem of rainfall return period in the field of geological hazards, and realizes the visualization of local stability in the landslide area, which can promote to enhance the ability of preventing and controlling landslide hazards.
( Linwei Li ),( Yiping Wu ),( Fasheng Miao ),( Yang Xue ),( Longfei Zhang ),( Kang Liao ),( Weifu Teng ),( Honglian Shi ) 대한지질공학회 2019 대한지질공학회 학술발표회논문집 Vol.2019 No.2
To overcome the drawbacks of previous displacement prediction models for step-like landslides, such as poor performance in predicting mutational displacement and unclear reliability of prediction results, this paper proposes a new hybrid method of landslide displacement prediction intervals. Firstly, the combination of SOM network and K-means clustering is implemented to divide the deformation states of step-like landslides into steady state and mutational state. Secondly, on the basis of expanding the mutational state samples through the comprehensive application of the engineering geology analogy method and the adaptive synthetic sampling algorithm, the random forest algorithm is used to establish an ensemble classifier for recognizing the landslide deformation states automatically. Finally, based on the Bootstrap-KELM-BPNN model, an interval prediction framework considering the dynamic switching of landslide deformation states is constructed to realize the dynamic prediction of landslide displacement. Taking Baishuihe landslide, a typical step-like landslide in the Three Gorges Reservoir Area, as an example, the dataset of XD01 monitoring point from June 2006 to December 2016 are explored to verify the effectiveness, accuracy and reliability of the proposed method.
( Kang Liao ),( Yiping Wu ),( Fasheng Miao ),( Linwei Li ),( Yang Xue ) 대한지질공학회 2019 대한지질공학회 학술발표회논문집 Vol.2019 No.2
Landslide displacement prediction is an essential topic in the landslide hazard research. Considering the characteristics of landslide deformation in the Three Gorges Reservoir Area of China, a step-like displacement prediction model based on Kernel Extreme Learning Machine with Grey Wolf Optimization (GWO-KELM) is proposed to predict the Baishuihe Landslide. First, the cumulative displacement is decomposed into the trend displacement and the periodic displacement by time series. Second, the trend displacement is predicted by a cubic polynomial model and the periodic displacement is predicted by the proposed model after statistically analyzing the displacement data. A hybrid model is then established for the prediction of landslide displacement. In addition, the performance of the hybrid model is compared with that of the Extreme Learning Machine with Grey Wolf Optimization (GWO-ELM), Support Vector Machine with Grey Wolf Optimization (GWO-SVM), and Extreme Learning Machine (ELM). The results show that the proposed hybrid model outperforms the other models, and the GWO-KELM model achieves a great ability in predicting the landslide displacement with a step-like behavior.