【摘要】：我國西南地區(qū)地質條件復雜，巖溶廣布，增加了施工過程中風險的難以預見性。傳統(tǒng)的風險分析方法主要是依靠專家的主觀經驗，，具有較大的局限性。本文引入了貝葉斯網絡理論，建立了貝葉斯網絡模型對巖溶隧道圍巖級別和圍巖穩(wěn)定性進行探討與分析,并得到了TSP系統(tǒng)和監(jiān)控量測的證明 本文以云桂高鐵倮得邑隧道為工程背景，針對巖溶隧道圍巖級別和圍巖穩(wěn)定性進行分析，主要研究內容包含以下幾個方面： 1.本文對國內外隧道工程風險管理和貝葉斯網絡理論發(fā)展與研究現(xiàn)狀進行了較詳細的論述。 2.本文詳細探討了風險分析與貝葉斯網絡的基本理論。闡述了風險分析的內容、步驟和方法，比較分析說明了貝葉斯后驗概率法的自身優(yōu)點。重點探討了貝葉斯網絡理論、貝葉斯網絡模型及其概率推理和簡化，并詳細介紹貝葉斯網絡模型仿真軟件Netica。 3.本文通過對圍巖分級綜合評判方法的探討，最終確定了巖溶隧道圍巖分級的基本指標及各指標的關系。并最終構建了巖溶隧道圍巖分級的貝葉斯網絡模型。利用模型仿真軟件Netica的后驗概率推理、最大可能說明以及敏感性分析三種功能，推導出巖溶對圍巖級別的影響最大。此外，在圍巖級別預測中，TSP超前地質預報證明了貝葉斯網絡模型的可行性。 4.本文通過對工程地質的自然因素和工程活動的人為因素的探討，確定巖溶隧道圍巖穩(wěn)定性影響指標。結合工程實際情況著重選取了拱效應、巖溶發(fā)育、開挖斷面、開挖擾動、支護強度和支護時機指標作為貝葉斯網絡節(jié)點。依據(jù)巖溶隧道風險分析的貝葉斯網絡模型，以IV級圍巖的穩(wěn)定性作為研究目標分析自然因素和人為因素對巖溶隧道圍巖穩(wěn)定性的影響。研究表明，巖溶發(fā)育情況對巖溶隧道圍巖穩(wěn)定性有著至關重要的影響。并通過工程現(xiàn)場監(jiān)控量測證明了貝葉斯網絡模型應對風險的適用性。
[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.
【學位授予單位】：湖南科技大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：U452.11

【相似文獻】

相關期刊論文 前10條

1 卓越;鄒

本文編號：2136042