Classification and susceptibility assessment of debris flow based on a semi-quantitative method combining of the fuzzy C-means algorithm, factor analysis and efficacy coefficient

Abstract

The existence of debris flows not only destroys the facilities, but also seriously threatens 9 human lives, especially in scenic areas. Therefore, the classification and susceptibility analysis of 10 debris flow are particularly important. In this paper, 21 debris flow catchments located in Huangsongyu 11 town ship, Pinggu District of Beijing, China were investigated. Besides field investigation, geographic 12 information system, global positioning system and remote sensing technology were applied to 13 determine the characteristics of debris flows. This article introduced clustering validity index to 14 determine the clustering number, and the fuzzy C-means algorithm and factor analysis method were 15 combined to classify 21 debris flow catchments in the study area. The results were divided into four 16 types: scale-topography-human activity closely related, topography-human activity-matter source 17 closely related, scale-matter source-geology closely related and topography-scale-matter source-human 18 activity closely related debris flow. And 9 major factors screened from the classification result were 19 selected for susceptibility analysis, using both the efficacy coefficient method and the combination 20 weighting. Susceptibility results showed that the susceptibility of 2 debris flows catchments were high, 216 were moderate, and 13 were low. The assessment results were consistent with the field investigation. 22Finally, a comprehensive assessment including classification and susceptibility evaluation of debris 23 flow was obtained, which was useful for risk mitigation and land use planning in the study area, and 24 provided reference for the research on related issues in other areas. 26Efficacy coefficient method 28 hundreds of people lose their lives, resulting in irreparable losses (Kang et al., 2004). 36Debris flow susceptibility analysis (DFS), which expresses the likelihood of a debris flow 37 occurring in an area with respect to its geomorphologic characteristics (Blais et al., 2016), is very 38 important to mitigate, evaluate and control debris flow disasters (Chiou et al., 2015). Physical, 39 empirical, and statistical approaches are used to analyze debris flow, which expresses the presumption 40 of a debris flow occurring in an area with respect to its geomorphologic characteristics (Blais et al., 41 2016). Physical-based approaches (Carrara et al., 2008; Burton and Bathurst, 1998) are more applicable 42to analyze physical and mechanical factors in independent catchments. Empirical model belongs to 43 qualitative evaluation and is too subjective to be convinced. Statistical analyses which are usually 44 applied in the research of regional debris flow, belongs to quantitative evaluation and depends on the 45 completeness and accuracy of data. For a study area with a limited number of debris flows, a 46 semi-quantitative evaluation method is more appropriate. This analysis includes the extraction of 47Proterozoic sedimentary strata and Quaternary sediments. The main lithology of the Archean age ...

The Reform of Engineering Professional Online Education Courses by Artificial Intelligence and Wireless Network Technology in the Context of Engineering Certification

“…Set l as the fuzzy index, then d ij is the Euclidean distance between the i-th user and the j-th center of the cluster. To calculate the objective function of the FCM clustering algorithm [28][29][30], according to n users and c user clusters, the data of matrix X is used and calculated, as shown in the following equation:…”