IMHE OpenIR  > 山地灾害与地表过程重点实验室
川藏铁路沿线大型滑坡早期判识研究
Alternative TitleStudy on Early Identification of Large-scale Landslide along Sichuan-Tibet Railway
边江豪
Subtype硕士
Thesis Advisor李秀珍
2018
Degree Grantor中国科学院大学
Place of Conferral北京
Degree Discipline建筑与土木工程
Keyword川藏铁路 大型滑坡 判识指标 支持向量机 智能判识
Abstract川藏铁路是国家重大基础设施建设项目,保障铁路顺利的建设以及未来的安全运营是重中之重。然而铁路沿线区域地势起伏大、构造运动强、新构造运动活跃、强震频率高、河流下切强,断裂构造和节理裂隙发育,陡峭斜坡与深切河谷地貌广泛分布,在各种内外营力共同作用下,使得成为我国山地灾害最发育、危害最严重的地区之一。尤其是大型滑坡发育,造成的危害往往更加严重,对拟建铁路是巨大威胁。因此,开展滑坡早期判识研究,对保障川藏铁路科学选定线以及川藏铁路的安全施工和运营具有重要的现实意义。本论文以川藏铁路沿线发育多、危害严重的大型滑坡作为主要研究对象,结合STS项目的相关调查和研究成果,在对滑坡发育环境条件进行分析的基础上,对川藏铁路沿线大型滑坡类型、特征及分布规律和主控因素等进行了系统分析。在此基础上,梳理了大型滑坡的判识途径和方法,建立了大型滑坡的判识指标体系,并建立了基于支持向量机方法的大型滑坡智能判识方法,旨在对川藏铁路沿线的大型滑坡及其稳定性进行早期判识,以期为川藏铁路选定线及安全施工运营提供依据。本文在对滑坡发育环境条件进行分析的基础上,对铁路沿线大型滑坡类型、特征、分布规律等进行了详细分析。川藏铁路沿线共发育滑坡崩塌灾害747处,其中大型滑坡(含特大型滑坡)共147处;大型滑坡类型包括硬岩类滑坡、软岩类滑坡、松散堆积层滑坡三种。大型滑坡在白玉至江达段分布密度最大,朗县至加查段和昌都到八宿段次之。利用主成分分析方法研究得出坡度、地层岩性、斜坡高差、坡体结构、断裂带等是大型滑坡形成和孕育的主控因素。结合已有文献资料,梳理了滑坡早期判识的途径及技术方法,总结了滑坡宏观定性判识指标,重点对铁路沿线发育较多的软岩类滑坡的早期判识指标体系进行分析。基于前述滑坡分布规律及主控因素的分析,建立了基于主控因素分析的滑坡半定性半定量判识指标,同时建立了滑坡的早期稳定性综合判识指标。采用人工智能的支持向量机方法,建立了判断是否为滑坡的智能判识模型和判识稳定性的多分类支持向量机智能模型。将建立的智能判识模型应用在实例中,分别进行滑坡判识与早期稳定性判识。
Other AbstractSichuan-tibet railway is a major national infrastructure construction project, and it is of great importance to guarantee the smooth construction of railway and the safe operation in the future. However, The area along the railway has big ups and downs, strong tectonic movement. new tectonic movement is active, strong earthquake frequency is high, the river incised, fracture and joint fissure have developed, and steep slope and deep valley landforms are widely distributed. Under the combination of various internal and external forces, it has become one of the most developed and most harmful areas in China。In particular, large landslides are so widespread that the damage is often more serious, it is a great threat to the proposed railway. Therefore, Therefore, the study on the early identification of landslides has important practical significance for ensuring the scientific selected lines of the Sichuan-Tibet Railway and the safe construction and operation of the Sichuan-Tibet. This dissertation focuses on the large-scale landslides that are heavily developed and endangered along the Sichuan-Tibet Railway. Combining the relevant investigations and research results of the STS project and based on the analysis of the landslide development environmental conditions, the landslide type along the Sichuan-Tibet Railway is considered as a major landslide type. Based on the analysis of the landslide development environment conditions, the types, characteristics, distribution and main controlling factors of large-scale landslides along the Sichuan-Tibet railway are systematically analyzed. On this basis, the ways and methods of identifying large-scale landslides were combed, The recognition index system of large landslide is established, and the intelligent recognition method of large landslide based on support vector machine is established, The purpose of this paper is to identify the large landslide and its stability along the Sichuan-Tibet railway. This article is based on the analysis of landslide development environmental conditions, the types, characteristics and distribution along the Sichuan-Tibet railway are systematically analyzed. This paper is concluded that there are 747 landslides and collapses along the railway line, including 147 large landslides (super large landslides). Large landslide types include hard rock landslide, soft rock landslide and loose accumulation layer landslide. The distribution density of large landslide is the highest in Baiyu to Jiangda section of Sichuan-Tibet railway, followed by Langxian to Jiacha section and Changdu to Basu section. By using principal component analysis method, it is concluded that slope, stratigraphic lithology, slope height difference, slope structure and fault zone are the main controlling factors for the formation and breeding of large-scale landslides.Combined with the existing literature, this paper combs the ways and technical methods of early identification of landslide, summarizes the macroscopic qualitative identification index of landslide, and analyzes the early identification index system of soft rock landslide which is more developed along railway line. Based on the analysis of the distribution law of the landslides and the main control factors, a semi quantitative semi quantitative identification index for landslide based on the analysis of the main control factors is established, and a comprehensive evaluation index for the early stability of the landslide is established.Using the support vector machine (SVM) method of artificial intelligence, the intelligent recognition model and the multi-classification support vector machine intelligent model are established to judge whether it is a landslide or not. The established intelligent recognition model is applied to an example to identify landslide and early stability respectively. 
Pages107
Language中文
Document Type学位论文
Identifierhttp://ir.imde.ac.cn/handle/131551/24767
Collection山地灾害与地表过程重点实验室
Affiliation中国科学院成都山地灾害与环境研究所
First Author Affilication中国科学院水利部成都山地灾害与环境研究所
Recommended Citation
GB/T 7714
边江豪. 川藏铁路沿线大型滑坡早期判识研究[D]. 北京. 中国科学院大学,2018.
Files in This Item:
File Name/Size DocType Version Access License
川藏铁路沿线大型滑坡早期判识研究.pdf(6482KB)学位论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[边江豪]'s Articles
Baidu academic
Similar articles in Baidu academic
[边江豪]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[边江豪]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 川藏铁路沿线大型滑坡早期判识研究.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.