基于機(jī)器學(xué)習(xí)技術(shù)的隧道掘進(jìn)機(jī)性狀的預(yù)測(cè)模型研究
發(fā)布時(shí)間：2021-07-11 16:14
盾構(gòu)隧道掘進(jìn)機(jī)器性能和刀具磨損的預(yù)測(cè)是一個(gè)非線性和多變量的復(fù)雜問題。為解決這個(gè)問題,本研究旨在:i)建立確定隧道掘進(jìn)機(jī)性能的智能分析框架,ii)預(yù)測(cè)隧道掘進(jìn)過程中的機(jī)器性能（即盾構(gòu)掘進(jìn)效率和盾構(gòu)切入速率）,iii)建立預(yù)測(cè)盾構(gòu)刀盤壽命的智能化統(tǒng)計(jì)模型,iv)分析隧道施工過程中每個(gè)參數(shù)的作用效應(yīng),特征和影響因素。研究過程中,應(yīng)用統(tǒng)計(jì)分析,機(jī)器學(xué)習(xí)技術(shù),智能分析和現(xiàn)場(chǎng)實(shí)測(cè)數(shù)據(jù)驗(yàn)證等手段研究這一系列問題。首先,通過確定隧道掘進(jìn)過程性能預(yù)測(cè)中最有效的參數(shù),提出盾構(gòu)掘進(jìn)效率和盾構(gòu)切入速率的新預(yù)測(cè)模型;然后,建立盾構(gòu)刀盤壽命的智能化新模型來預(yù)測(cè)刀盤壽命;為了獲得更為可靠的施工操作,基于地層力學(xué)參數(shù)與盾構(gòu)施工參數(shù)兩個(gè)方面,提出一種智能分析方法;最后,討論分析盾構(gòu)刀盤壽命預(yù)測(cè)中最重要影響參數(shù)的作用,確定預(yù)測(cè)模型。研究的創(chuàng)新成果總結(jié)如下:（1）提出了新的機(jī)器學(xué)習(xí)模型預(yù)測(cè)盾構(gòu)機(jī)的掘進(jìn)效率提出的機(jī)器學(xué)習(xí)模型集成了改進(jìn)的粒子群優(yōu)化（PSO）合法自適應(yīng)神經(jīng)模糊推理系統(tǒng)（ANFIS）在一起。提出的改進(jìn)模型組合了基于模糊規(guī)則的系統(tǒng)和PSO算法,可以同時(shí)調(diào)整先行變量和后續(xù)變量。提出的模型與當(dāng)前廣泛使用模型,如神經(jīng)網(wǎng)絡(luò)...
【文章來源】：上海交通大學(xué)上海市 211工程院校 985工程院校 教育部直屬院校
【文章頁(yè)數(shù)】：212 頁(yè)
【學(xué)位級(jí)別】：博士
【文章目錄】：
Abstract
摘要
List of acronyms and abbreviations
Chapter1 Introduction
1.1 Background and motivation
1.2 Role of models in tunneling
1.3 Definition of the problem
1.4 Objectives of this study
1.5 Research strategy/design
1.6 Structure of this dissertation
Chapter2 Literature review
2.1 Introduction
2.2 TBM Tunneling
2.2.1 Working principle
2.2.2 Earth pressure balance(EPB)tunnel boring machine
2.3 Parameters influencing excavation performance and tool wear
2.3.1 Face pressure
2.3.2 Screw conveyor
2.3.3 Thrust and torque of cutter wheel
2.3.4 Soil conditioning agent
2.3.5 Cutter wheel rotation speed
2.3.6 Penetration rate,utilization factor,and advance rate
2.4 Current state of disc cutter design and development direction
2.4.1 Disc cutter life(Hf)prediction model
2.5 TBM prediction models
2.5.1 Artificial intelligent techniques
2.5.2 Optimization techniques
2.5.3 Evaluation TBM performance through AI techniques
2.6 Summary
Chapter3 Data-driven framework for improving shield performance
3.1 Introduction
3.2 Visual analysis of the data
3.2.1 Statistical modeling
3.2.2 Principal component analysis
3.2.3 Simple regression analysis
3.2.4 Non-linear multiple regression analysis
3.3 Advance rate prediction through neural network model
3.3.1 Neural network architecture selection
3.3.2 Analysis of neural network
3.4 Advance rate prediction through fuzzy logic model
3.4.1 Fuzzification part
3.4.2 Knowledge base
3.4.3 Fuzzy inference system(FIS)
3.4.4 Defuzzification process
3.4.5 Analysis of fuzzy logic
3.5 Advance rate prediction through ANFIS techniques
3.5.1 ANFIS Architecture
3.5.2 Hybrid learning algorithm
3.5.3 Structure identification methods
3.6 Summary
Chapter4 Machine performance using optimization models
4.1 Introduction
4.2 Estimating TBM performance
4.3 Proposed technique for advance rate prediction
4.3.1 Original PSO algorithm
4.3.2 Improvement of inertia weight
4.3.3 Improvement of constriction factor
4.3.4 Synchronously inertia weight and constriction factor
4.3.5 Hybrid improved IPSO-ANFIS model
4.3.6 Model evaluations
4.3.7 Comparison of IPSO-ANFIS model with other techniques
4.4 Proposed technique for penetration rate prediction
4.4.1 Genetic algorithm
4.4.2 Improving ANFIS using GA model
4.4.3 Multi-objective fitness function
4.4.4 Model evaluation
4.4.5 Comparison of multi-objective optimization model with other technique
4.5 Discussion
4.6 Summary
Chapter5 Intelligent approach to estimate disc cutter life
5.1 Introduction
5.2 Visualization to estimate the consumption of disc cutter
5.3 Developing model for estimating disc cutter life
5.3.1 Statistical analysis
5.3.2 Simple regression analysis
5.3.3 Non-linear multiple regression analysis
5.4 An intelligence technique
5.4.1 Group method of data handling polynomial neural network
5.4.2 Hybrid GMDH-GA technique
5.4.3 Evaluation methodology of cutter life using optimized GMDH-GA
5.4.4 Model validation
5.5 Analyze the efficiency of parameters to predict cutter life
5.6 Summary
Chapter6 Case studies:prediction of tunnel performance
6.1 Introduction
6.2 Guangzhou Metro Line no.9(Case study)
6.2.1 Project description
6.2.2 Geological conditions
6.2.3 Rock percentage encountered the tunnel face
6.2.4 Cutter wear and its effect on shield advancement rate
6.2.5 Effect of TBM field database on advance rate
6.3 Guangzhou-Shenzhen intercity railway project
6.3.1 Project description
6.3.2 Geological conditions
6.3.3 Disc cutter consumption
6.3.4 Analysis of shield parameters
6.4 Discussion
6.4.1 Visualization of the evolving models for TBM performance
6.4.2 Visualization of the evolving models for disc cutter life
6.5 Summary
Chapter7 Concluding remarks
7.1 A brief summary
7.2 Limitations
7.3 Perspective
Appendix A
Appendix B
References
Acknowledgements
Curriculum vitae
Publications during my Ph D study
【參考文獻(xiàn)】：
期刊論文
[1]The comparative analysis of rocks’ resistance to forward-slanting disc cutters and traditionally installed disc cutters[J]. Zhao-Huang Zhang,Sun Fei,Meng Liang. Acta Mechanica Sinica. 2016(04)
[2]巨斑狀花崗巖條件下TBM大直徑盤形滾刀磨耗規(guī)律[J]. 杜立杰,紀(jì)珊珊,左立富,孔海峽,許金林,杜彥良. 煤炭學(xué)報(bào). 2015(12)
[3]上軟下硬地層碴土改良試驗(yàn)及應(yīng)用研究[J]. 葉新宇,王樹英,肖超,陽(yáng)軍生,周純擇. 現(xiàn)代隧道技術(shù). 2015(06)
[4]TBM盤形滾刀在山嶺隧道掘進(jìn)過程中的磨損研究[J]. 趙戰(zhàn)欣. 地下空間與工程學(xué)報(bào). 2015(S1)
[5]慣性權(quán)值對(duì)粒子群算法收斂性的影響及改進(jìn)[J]. 黃翀鵬,熊偉麗,徐保國(guó). 計(jì)算機(jī)工程. 2008(12)
[6]粒子群優(yōu)化算法的收斂性分析及其混沌改進(jìn)算法[J]. 劉洪波,王秀坤,譚國(guó)真. 控制與決策. 2006(06)
[7]盤形滾刀的使用與研究（1）——TB880E型掘進(jìn)機(jī)在秦嶺隧道施工中的應(yīng)用[J]. 萬(wàn)治昌,沙明元,周雁領(lǐng). 現(xiàn)代隧道技術(shù). 2002(05)
[8]隧道掘進(jìn)機(jī)在中國(guó)地下工程中應(yīng)用現(xiàn)狀及前景展望[J]. 錢七虎,李朝甫,傅德明. 地下空間. 2002(01)
[9]基于粒子群優(yōu)化的文檔聚類算法[J]. 魏建香,孫越泓,蘇新寧. 情報(bào)學(xué)報(bào). 2010 (03)
本文編號(hào)：3278422
【文章來源】：上海交通大學(xué)上海市 211工程院校 985工程院校 教育部直屬院校
【文章頁(yè)數(shù)】：212 頁(yè)
【學(xué)位級(jí)別】：博士
【文章目錄】：
Abstract
摘要
List of acronyms and abbreviations
Chapter1 Introduction
1.1 Background and motivation
1.2 Role of models in tunneling
1.3 Definition of the problem
1.4 Objectives of this study
1.5 Research strategy/design
1.6 Structure of this dissertation
Chapter2 Literature review
2.1 Introduction
2.2 TBM Tunneling
2.2.1 Working principle
2.2.2 Earth pressure balance(EPB)tunnel boring machine
2.3 Parameters influencing excavation performance and tool wear
2.3.1 Face pressure
2.3.2 Screw conveyor
2.3.3 Thrust and torque of cutter wheel
2.3.4 Soil conditioning agent
2.3.5 Cutter wheel rotation speed
2.3.6 Penetration rate,utilization factor,and advance rate
2.4 Current state of disc cutter design and development direction
2.4.1 Disc cutter life(Hf)prediction model
2.5 TBM prediction models
2.5.1 Artificial intelligent techniques
2.5.2 Optimization techniques
2.5.3 Evaluation TBM performance through AI techniques
2.6 Summary
Chapter3 Data-driven framework for improving shield performance
3.1 Introduction
3.2 Visual analysis of the data
3.2.1 Statistical modeling
3.2.2 Principal component analysis
3.2.3 Simple regression analysis
3.2.4 Non-linear multiple regression analysis
3.3 Advance rate prediction through neural network model
3.3.1 Neural network architecture selection
3.3.2 Analysis of neural network
3.4 Advance rate prediction through fuzzy logic model
3.4.1 Fuzzification part
3.4.2 Knowledge base
3.4.3 Fuzzy inference system(FIS)
3.4.4 Defuzzification process
3.4.5 Analysis of fuzzy logic
3.5 Advance rate prediction through ANFIS techniques
3.5.1 ANFIS Architecture
3.5.2 Hybrid learning algorithm
3.5.3 Structure identification methods
3.6 Summary
Chapter4 Machine performance using optimization models
4.1 Introduction
4.2 Estimating TBM performance
4.3 Proposed technique for advance rate prediction
4.3.1 Original PSO algorithm
4.3.2 Improvement of inertia weight
4.3.3 Improvement of constriction factor
4.3.4 Synchronously inertia weight and constriction factor
4.3.5 Hybrid improved IPSO-ANFIS model
4.3.6 Model evaluations
4.3.7 Comparison of IPSO-ANFIS model with other techniques
4.4 Proposed technique for penetration rate prediction
4.4.1 Genetic algorithm
4.4.2 Improving ANFIS using GA model
4.4.3 Multi-objective fitness function
4.4.4 Model evaluation
4.4.5 Comparison of multi-objective optimization model with other technique
4.5 Discussion
4.6 Summary
Chapter5 Intelligent approach to estimate disc cutter life
5.1 Introduction
5.2 Visualization to estimate the consumption of disc cutter
5.3 Developing model for estimating disc cutter life
5.3.1 Statistical analysis
5.3.2 Simple regression analysis
5.3.3 Non-linear multiple regression analysis
5.4 An intelligence technique
5.4.1 Group method of data handling polynomial neural network
5.4.2 Hybrid GMDH-GA technique
5.4.3 Evaluation methodology of cutter life using optimized GMDH-GA
5.4.4 Model validation
5.5 Analyze the efficiency of parameters to predict cutter life
5.6 Summary
Chapter6 Case studies:prediction of tunnel performance
6.1 Introduction
6.2 Guangzhou Metro Line no.9(Case study)
6.2.1 Project description
6.2.2 Geological conditions
6.2.3 Rock percentage encountered the tunnel face
6.2.4 Cutter wear and its effect on shield advancement rate
6.2.5 Effect of TBM field database on advance rate
6.3 Guangzhou-Shenzhen intercity railway project
6.3.1 Project description
6.3.2 Geological conditions
6.3.3 Disc cutter consumption
6.3.4 Analysis of shield parameters
6.4 Discussion
6.4.1 Visualization of the evolving models for TBM performance
6.4.2 Visualization of the evolving models for disc cutter life
6.5 Summary
Chapter7 Concluding remarks
7.1 A brief summary
7.2 Limitations
7.3 Perspective
Appendix A
Appendix B
References
Acknowledgements
Curriculum vitae
Publications during my Ph D study
【參考文獻(xiàn)】：
期刊論文
[1]The comparative analysis of rocks’ resistance to forward-slanting disc cutters and traditionally installed disc cutters[J]. Zhao-Huang Zhang,Sun Fei,Meng Liang. Acta Mechanica Sinica. 2016(04)
[2]巨斑狀花崗巖條件下TBM大直徑盤形滾刀磨耗規(guī)律[J]. 杜立杰,紀(jì)珊珊,左立富,孔海峽,許金林,杜彥良. 煤炭學(xué)報(bào). 2015(12)
[3]上軟下硬地層碴土改良試驗(yàn)及應(yīng)用研究[J]. 葉新宇,王樹英,肖超,陽(yáng)軍生,周純擇. 現(xiàn)代隧道技術(shù). 2015(06)
[4]TBM盤形滾刀在山嶺隧道掘進(jìn)過程中的磨損研究[J]. 趙戰(zhàn)欣. 地下空間與工程學(xué)報(bào). 2015(S1)
[5]慣性權(quán)值對(duì)粒子群算法收斂性的影響及改進(jìn)[J]. 黃翀鵬,熊偉麗,徐保國(guó). 計(jì)算機(jī)工程. 2008(12)
[6]粒子群優(yōu)化算法的收斂性分析及其混沌改進(jìn)算法[J]. 劉洪波,王秀坤,譚國(guó)真. 控制與決策. 2006(06)
[7]盤形滾刀的使用與研究（1）——TB880E型掘進(jìn)機(jī)在秦嶺隧道施工中的應(yīng)用[J]. 萬(wàn)治昌,沙明元,周雁領(lǐng). 現(xiàn)代隧道技術(shù). 2002(05)
[8]隧道掘進(jìn)機(jī)在中國(guó)地下工程中應(yīng)用現(xiàn)狀及前景展望[J]. 錢七虎,李朝甫,傅德明. 地下空間. 2002(01)
[9]基于粒子群優(yōu)化的文檔聚類算法[J]. 魏建香,孫越泓,蘇新寧. 情報(bào)學(xué)報(bào). 2010 (03)
本文編號(hào)：3278422
本文鏈接：http://sikaile.net/kejilunwen/daoluqiaoliang/3278422.html
最近更新
教材專著
