基于贝叶斯网络的高铁岩溶隧道风险分析
发布时间：2018-07-21 16:01
【摘要】：我国西南地区地质条件复杂，岩溶广布，增加了施工过程中风险的难以预见性。传统的风险分析方法主要是依靠专家的主观经验，，具有较大的局限性。本文引入了贝叶斯网络理论，建立了贝叶斯网络模型对岩溶隧道围岩级别和围岩稳定性进行探讨与分析,并得到了TSP系统和监控量测的证明 本文以云桂高铁倮得邑隧道为工程背景，针对岩溶隧道围岩级别和围岩稳定性进行分析，主要研究内容包含以下几个方面： 1.本文对国内外隧道工程风险管理和贝叶斯网络理论发展与研究现状进行了较详细的论述。 2.本文详细探讨了风险分析与贝叶斯网络的基本理论。阐述了风险分析的内容、步骤和方法，比较分析说明了贝叶斯后验概率法的自身优点。重点探讨了贝叶斯网络理论、贝叶斯网络模型及其概率推理和简化，并详细介绍贝叶斯网络模型仿真软件Netica。 3.本文通过对围岩分级综合评判方法的探讨，最终确定了岩溶隧道围岩分级的基本指标及各指标的关系。并最终构建了岩溶隧道围岩分级的贝叶斯网络模型。利用模型仿真软件Netica的后验概率推理、最大可能说明以及敏感性分析三种功能，推导出岩溶对围岩级别的影响最大。此外，在围岩级别预测中，TSP超前地质预报证明了贝叶斯网络模型的可行性。 4.本文通过对工程地质的自然因素和工程活动的人为因素的探讨，确定岩溶隧道围岩稳定性影响指标。结合工程实际情况着重选取了拱效应、岩溶发育、开挖断面、开挖扰动、支护强度和支护时机指标作为贝叶斯网络节点。依据岩溶隧道风险分析的贝叶斯网络模型，以IV级围岩的稳定性作为研究目标分析自然因素和人为因素对岩溶隧道围岩稳定性的影响。研究表明，岩溶发育情况对岩溶隧道围岩稳定性有着至关重要的影响。并通过工程现场监控量测证明了贝叶斯网络模型应对风险的适用性。
[Abstract]:The geological conditions in southwest China are complicated and the karst is widespread, which increases the unpredictable risk in the construction process. The traditional risk analysis method is mainly based on the subjective experience of experts and has great limitations. In this paper, the Bayesian network theory is introduced, and the Bayesian network model is established to discuss and analyze the surrounding rock grade and stability of karst tunnel. It is proved by tsp system and monitoring measurement that this paper takes Luodeyi tunnel of Yun-Gui high-speed railway as the engineering background and analyzes the surrounding rock grade and surrounding rock stability of karst tunnel. The main research contents include the following aspects: 1. In this paper, the development and research status of risk management and Bayesian network theory in tunnel engineering at home and abroad are discussed in detail. 2. In this paper, the basic theory of risk analysis and Bayesian network is discussed in detail. The contents, steps and methods of risk analysis are described, and the advantages of Bayesian posteriori probability method are illustrated. The Bayesian network theory, Bayesian network model and its probabilistic reasoning and simplification are discussed in detail, and the simulation software Netica.3. of Bayesian network model is introduced in detail. This paper discusses the comprehensive evaluation method of surrounding rock classification, and finally determines the basic index of surrounding rock classification of karst tunnel and the relation of each index. Finally, the Bayesian network model of surrounding rock classification of karst tunnel is constructed. By using the posteriori probability reasoning of Netica, maximum possibility explanation and sensitivity analysis, it is deduced that karst has the greatest influence on the surrounding rock level. In addition, the advance geological prediction of tsp in the prediction of surrounding rock level proves the feasibility of Bayesian network model. 4. Based on the discussion of the natural factors of engineering geology and the man-made factors of engineering activities, the influence indexes of surrounding rock stability of karst tunnels are determined in this paper. Combined with the actual situation of the project, the key points of Bayesian network are the arch effect, karst development, excavation section, excavation disturbance, support strength and supporting timing index. Based on Bayesian network model of risk analysis of karst tunnel, the influence of natural and human factors on the stability of karst tunnel surrounding rock is analyzed with the stability of grade IV surrounding rock as the research objective. The study shows that karst development has an important effect on the stability of karst tunnel surrounding rock. The applicability of Bayesian network model to risk is proved by field monitoring and measurement.
【学位授予单位】：湖南科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U452.11
[Abstract]:The geological conditions in southwest China are complicated and the karst is widespread, which increases the unpredictable risk in the construction process. The traditional risk analysis method is mainly based on the subjective experience of experts and has great limitations. In this paper, the Bayesian network theory is introduced, and the Bayesian network model is established to discuss and analyze the surrounding rock grade and stability of karst tunnel. It is proved by tsp system and monitoring measurement that this paper takes Luodeyi tunnel of Yun-Gui high-speed railway as the engineering background and analyzes the surrounding rock grade and surrounding rock stability of karst tunnel. The main research contents include the following aspects: 1. In this paper, the development and research status of risk management and Bayesian network theory in tunnel engineering at home and abroad are discussed in detail. 2. In this paper, the basic theory of risk analysis and Bayesian network is discussed in detail. The contents, steps and methods of risk analysis are described, and the advantages of Bayesian posteriori probability method are illustrated. The Bayesian network theory, Bayesian network model and its probabilistic reasoning and simplification are discussed in detail, and the simulation software Netica.3. of Bayesian network model is introduced in detail. This paper discusses the comprehensive evaluation method of surrounding rock classification, and finally determines the basic index of surrounding rock classification of karst tunnel and the relation of each index. Finally, the Bayesian network model of surrounding rock classification of karst tunnel is constructed. By using the posteriori probability reasoning of Netica, maximum possibility explanation and sensitivity analysis, it is deduced that karst has the greatest influence on the surrounding rock level. In addition, the advance geological prediction of tsp in the prediction of surrounding rock level proves the feasibility of Bayesian network model. 4. Based on the discussion of the natural factors of engineering geology and the man-made factors of engineering activities, the influence indexes of surrounding rock stability of karst tunnels are determined in this paper. Combined with the actual situation of the project, the key points of Bayesian network are the arch effect, karst development, excavation section, excavation disturbance, support strength and supporting timing index. Based on Bayesian network model of risk analysis of karst tunnel, the influence of natural and human factors on the stability of karst tunnel surrounding rock is analyzed with the stability of grade IV surrounding rock as the research objective. The study shows that karst development has an important effect on the stability of karst tunnel surrounding rock. The applicability of Bayesian network model to risk is proved by field monitoring and measurement.
【学位授予单位】：湖南科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U452.11
【相似文献】
相关期刊论文 前10条
1 卓越;邹
本文编号：2136042
本文链接：https://www.wllwen.com/kejilunwen/jiaotonggongchenglunwen/2136042.html
