【作者】 刘震；

【导师】 钟登华；

【作者基本信息】 天津大学 ， 水工结构工程， 2018， 博士

【摘要】 复杂深埋引水隧洞群洞室布置纵横交错,施工进度计划和资源优化配置非常复杂,其施工过程是一个多层次、多工序的动态循环过程;此外,高地温围岩的传热影响导致隧洞内通风环境进一步恶化,劳动生产率降低,威胁人员生命安全。目前引水隧洞施工仿真研究中通风参数取值多依据工程经验或局限于单一通风方案的等温通风数值模拟;同时,现有隧洞施工仿真研究大多忽略了仿真参数的动态变化特征,已有的少数研究局限于采用贝叶斯参数更新技术,难以准确描述参数变化过程;因此,如何在通风参数的优化取值研究中考虑高地温围岩的传热影响和动态变化的通风需求,如何准确描述施工仿真参数随施工推进而动态演变的特征是复杂深埋引水隧洞群施工仿真优化领域亟待解决的重要课题。本文针对上述问题开展了深入的研究,并取得了如下创新性研究成果:(1)针对当前复杂深埋引水隧洞群施工仿真研究中缺乏综合考虑高地温围岩传热影响和通风方案动态变化的仿真通风参数科学取值的不足,以及参数更新方法难以准确描述施工参数动态演变特征的现状,提出了复杂深埋引水隧洞群施工仿真优化理论与方法。传统复杂深埋引水隧洞群施工仿真研究中通风参数取值多依据工程经验或局限于单一通风方案的等温通风数值模拟,缺乏综合考虑高地温围岩传热影响和通风方案动态变化的仿真通风参数科学取值的研究;此外,现有隧洞施工仿真研究大多忽略了仿真参数随施工推进的动态变化特征,已有少数研究均局限于采用贝叶斯参数更新技术,难以精确描述参数变化过程。针对上述问题,提出了复杂深埋引水隧洞群施工仿真优化理论与方法。首先,提出综合考虑通风参数优化取值和施工参数动态更新的施工仿真优化流程;其次,构建耦合CPM(Critical Path Method)模型和CYCLONE(Cycle Operation Network)模型的复杂深埋引水隧洞群施工仿真优化模型;最后,提出施工强度统计和资源均衡优化分析方法,为复杂深埋引水隧洞群施工组织设计和项目管理决策提供理论支持。(2)针对当前隧洞围岩传热数值模拟研究中大多将岩石导热系数当作围岩导热系数,缺乏考虑实际裂隙分布特征的隧洞围岩等效导热系数研究的现状,提出基于三维裂隙网络分形分析的复杂深埋引水隧洞群围岩等效导热系数分析方法。现有隧洞工程围岩传热数值模拟研究大多把岩石导热系数当作围岩导热系数,忽略了裂隙对围岩导热系数的重要影响;此外现有岩体等效导热系数研究局限于现场试验、二维传热模拟以及基于孔隙的随机建模模拟,缺乏考虑实际裂隙分布特征的隧洞围岩三维裂隙岩体等效导热系数研究。针对上述问题,提出了基于三维裂隙网络分形分析的复杂深埋引水隧洞群围岩等效导热系数分析方法。首先,利用复杂深埋引水隧洞群围岩揭露面裂隙编录数据,基于概率论和统计学理论,采用Monte Carlo随机模拟方法对裂隙的几何参数进行模拟,进而建立围岩三维随机裂隙网络模型;其次,基于分形理论采用立方体计盒维数法分析了三维随机裂隙网络模型的分形维数;最后,结合三维裂隙网络模型及其分形特征推导了围岩等效导热系数的数学表达式,为改进考虑围岩传热影响的复杂深埋引水隧洞群施工通风数学模型提供了理论依据。(3)针对现有引水隧洞施工通风模拟研究中匮乏综合考虑围岩热传导以及风流与壁面热交换耦合影响的复杂深埋引水隧洞群施工通风两相流紊流模型研究的现状,提出了考虑围岩传热影响的复杂深埋引水隧洞群施工通风两相流模拟方法。目前引水隧洞施工通风的两相流模拟研究主要集中在等温通风方面,此外现有深埋引水隧洞的非等温两相流模拟研究局限于考虑风流与壁面的热交换作用,忽略了高地温围岩传热对于施工通风的重要影响。针对上述问题,提出了考虑围岩传热影响的复杂深埋引水隧洞群施工通风两相流模拟方法。首先,建立综合考虑围岩内部热传导以及风流与壁面间热交换耦合影响的复杂深埋引水隧洞群施工通风两相流紊流模型,其中采用基于三维裂隙网络分形分析得出的围岩等效导热系数对围岩热传导模型进行改进;其次,针对复杂深埋引水隧洞群错综复杂的洞室布置和随施工推进动态变化的通风需求,确定合理的通风方式、通风量等模拟参数;再者,采用网格收敛指数(Grid Convergence Index,GCI)和变量变化百分比(Variate Percentage Change,VPC)对计算网格和时间步长的独立性进行分析,得出了合理的网格划分方案和时间步长取值;最后,基于现场实测数据,验证了数学模型的可靠性,为实现仿真通风参数的优化取值提供了理论基础。(4)针对目前隧洞施工仿真研究中大多忽略了施工参数的动态演变特征,且现有参数更新方法存在参数更新频率较低、结果难以准确描述仿真参数变化特征的现状,提出了基于自适应混沌差分进化支持向量机的复杂深埋引水隧洞群施工仿真参数动态更新方法。目前隧洞施工仿真研究大多忽略了施工参数的动态演变特征,已有的少数仿真参数更新研究局限于采用贝叶斯更新方法,均假设仿真参数取值服从正态分布,存在假设分布与真实分布不符、参数更新频率较低以及难以准确描述仿真参数非线性变化过程的不足。针对上述问题,提出了基于自适应混沌差分进化支持向量机的复杂深埋引水隧洞群施工仿真参数动态更新方法。首先,采用自适应缩放因子和混沌理论对差分进化算法进行改进,提出自适应混沌差分进化算法(Adaptive Chaos Differential Evolution,ACDE);其次,基于现场施工参数原始时间序列,采用ACDE对支持向量机(Support Vector Machine,SVM)进行参数寻优,进而构建基于自适应混沌差分进化算法支持向量机(ACDE-SVM)的施工仿真参数预测模型;再者,采用误差指标对模型性能进行评价,并与采用常规仿真方法以及贝叶斯更新方法的仿真结果进行对比,验证基于ACDE-SVM的仿真参数动态更新方法的优越性和适用性。(5)依托实际工程,基于综合考虑通风参数优化取值和仿真参数动态更新的复杂深埋引水隧洞群施工仿真优化理论与方法,实现了复杂深埋引水隧洞群施工仿真优化应用研究,验证了上述理论和方法的可行性。以某复杂深埋引水隧洞群工程为例,将上述提出的理论和方法应用到实际工程中。首先,基于复杂深埋引水隧洞群不同施工时期的通风需求,制定不同施工通风方案;其次,基于三维裂隙网络分形分析得出的围岩等效导热系数,采用考虑围岩传热的复杂深埋引水隧洞群施工通风两相流模拟方法对不同施工通风方案下的通风散烟降温过程进行模拟研究,进而分析风流场紊流结构特征,同时探讨隧洞内围岩热传导以及围岩与风流热交换规律、污染物迁移变化规律,并最终揭示不同施工通风方案下通风时间与开挖长度之间的内在联系,优化仿真通风参数取值;再者,依据现场施工参数原始时间序列,采用基于ACDE-SVM的施工仿真参数预测模型,进行了复杂深埋引水隧洞群施工仿真参数动态预测分析;最后,基于优化后的仿真通风参数和施工仿真参数预测时间序列,采用施工仿真优化模型,实现复杂深埋引水隧洞群施工仿真优化研究,为施工进度计划的安排以及现场施工工期的控制提供理论指导和技术支撑。更多还原

【Abstract】 The criss-cross cavern layout,complex construction schedule and resource optimization are characteristics of the complex deep buried diversion tunnel group,and its construction process is a multi-level and multi-process dynamic cycle.In addition,the heat transfer effect of high ground temperature surrounding rock leads to further deterioration of ventilation environment,lower labor productivity and endanger the life safety of workers.At present,the ventilation parameter values in the simulation study of the diversion tunnel are based on the engineering experience or the numerical simulation of isothermal ventilation confined to a single ventilation scheme.Meanwhile,most of the existing tunnel construction simulation ignores the dynamic change characteristics of the simulation parameters,and a few studies have been limited to the use of Bayesian updating methods,which is difficult to accurately describe process parameter change.Therefore,how to consider the influence of high ground temperature surrounding rock and the dynamic change of ventilation demand in the optimization of ventilation parameter as well as how to describe the dynamic evolution of construction simulation parameters with the construction propulsion are important subjects to be solved urgently in the field of construction simulation optimization of the complex deep buried diversion tunnel group.This paper makes an in-depth study on the problems mentioned above and obtains the following results:1.The influence of high ground temperature heat transfer and the dynamic change of ventilation scheme are not synthetically considered in determining the value of construction ventilation parameters in the existing research of complex deep buried diversion tunnel group construction simulation,and parameter updating methods are hard to accurately describe the dynamic evolution characteristics of construction parameters.Therefore,optimization theory and method of complex deep buried diversion tunnel group construction simulation are proposed.Ventilation parameter value selection in traditional complex deep buried diversion tunnel group construction simulation research is based on engineering experience or confined to isothermal ventilation simulation with single ventilation scheme,lacking overall consideration of influence of high ground temperature heat transfer and dynamic change of ventilation scheme.Besides,most of the existing tunnel construction simulation researches ignore the dynamic change features of simulation parameters with construction advance.And minority researches are limited to using Bayesian parameter updating technique,which is hard to precisely describe the parameter changing process.Therefore,optimization theory and method of complex deep buried diversion tunnel group construction simulation are proposed.Firstly,construction simulation optimize process in which the ventilation parameter value optimization selection and construction parameter dynamic update are synthetically considered.Then the complex deep buried diversion tunnel group construction simulation optimization model is established by couple CPM model and CYCLONE model.Finally,construction intensity statistics and resource balance optimization method is proposed to provide theoretical support of complex deep buried diversion tunnel group construction organization design and project management decision.2.Current studies on the numerical simulation of tunnel surrounding rock thermal conduction mostly regard the homogeneous rock thermal conductivity as the surrounding rock thermal conductivity,and there lacks the research on the effective thermal conductivity of tunnel surrounding rock that considering the rock mass as fractured media.This paper proposes a method of analyzing the surrounding rock effective thermal conductivity of complex deep buried diversion tunnel group based on fractal analysis of 3D fracture network.Current studies on the numerical simulation of tunnel surrounding rock thermal conduction mostly consider the homogeneous rock thermal conductivity as the surrounding rock thermal conductivity,and these studies focus on field testing,2D thermal conduction numerical simulation and porous media modeling and simulation,so there lacks the research on the effective thermal conductivity of tunnel surrounding rock considering the rock mass as fractured media.To solve these problems,this paper proposes a method of analyzing the surrounding rock effective thermal conductivity of complex deep buried diversion tunnel group based on fractal analysis of 3D fracture network.Firstly,the surrounding rock exposed fracture traces log data of actual complex deep buried diversion tunnel group is collected,and based on the probability and statistics theory,the geometric parameters of the fracture is estimated and the 3D fracture network is stochastically modeled with Monte Carlo method;Secondly,based on the fractal theory,the fractal dimension of the 3D random fracture network is calculated with cubic covering method;Finally,the mathematical expression of surrounding rock effective thermal conductivity is derived based on the fractal characteristic of the 3D fracture network,and this research provides a theoretical foundation for improving the mathematical model of complex deep buried diversion tunnel group construction ventilation simulation considering the influence of surrounding rock thermal conduction.3.In view of the actuality that synthetically consideration of heat conduction within surrounding rock and heat exchange coupling between wind flow and surrounding rock in the research of complex deep buried diversion tunnel group construction ventilation two-phase turbulent flow model is scarce,a complex deep buried diversion tunnel group construction ventilation two-phase flow simulation method considering the influence of surrounding rock heat transfer is put forward.At present the research of diversion tunnel construction ventilation two-phase flow simulation converges on isothermal ventilation,besides the existing non-isothermal two-phase flow simulation of deep buried diversion tunnel research is restricted to considering the heat exchange between wind flow and surrounding rock,and ignores the important influence of high ground temperature heat transfer on construction ventilation.Aiming at those problems,a complex deep buried diversion tunnel group construction ventilation two-phase flow simulation method considering the influence of surrounding rock heat transfer is proposed,and a complex deep buried diversion tunnel group construction ventilation two-phase turbulent flow model considering the influence of heat conduction within surrounding rock and heat exchange coupling between wind flow and surrounding rock is established.Firstly the surrounding rock heat transfer model is improved by equivalent thermal conductivity coefficient of surrounding rock obtained through analysis of three-dimensional fracture network fractal theory.Then reasonable simulation parameters,such as ventilation pattern,ventilation quantity and so on,are confirmed by considering the intricate cavity layout of complex deep buried diversion tunnel group and ventilation requirements changing with construction advance.Moreover,in order to obtain rational grid generation scheme and time step value,Grid Convergence Index(GCI)and Variate Percentage Change(VPC)are used to analyze the independence of grid and time step.Finally,the reliability of mathematic model is verified by field measured data to provide theoretical basis for getting optimal values of ventilation simulation parameters.4.Considering the problems that current tunnel construction simulation researches mostly ignore the dynamic evolution characteristics of construction parameters,the current parameter updating methods have low updating frequency,and the results cannot describe the change characteristics of the simulation parameter accurately.The dynamic updating method for simulation parameters of complex deep buried diversion tunnel group based on adaptive chaos differential evolution-support vector machine is proposed.In the current researches of tunnel construction simulation,most of them ignore the dynamic evolution characteristics of construction parameters,the existing few simulation parameter updating studies are limited to use the Bayesian updating method which assuming that the simulation parameter values obey the normal distribution,this leads to the inconsistency between the hypothesis distribution and the real distribution,the low frequency of parameter updating and the difficulty in accurately describing the nonlinear variation of simulation parameters.Aiming at the above problems,the dynamic updating method for simulation parameters of complex deep water diversion tunnel group based on adaptive chaos differential evolution-support vector machine is proposed.Firstly,adaptive scaling factor and Chaos theory are adopted to improve the Differential Evolution algorithm,and the Adaptive Chaos Differential Evolution algorithm(ACDE)is proposed;Secondly,based on the original time series of site construction parameters,the parameter optimization of support vector machine(SVM)is optimized by ACDE,and the construction simulation parameter prediction model based on adaptive chaotic differential evolution algorithm support vector machine(ACDE-SVM)is constructed;Then the error index is used to evaluate the performance of the model,and compared with the simulation results using the conventional simulation method and the Bayesian updating method,the superiority and applicability of the dynamic updating method of simulation parameters based on ACDE-SVM is verified.5.Based on practical engineering,the construction simulation optimization application of complex deep buried diversion tunnel group is realized by using the theory and method of the construction simulation optimization of complex deep buried diversion tunnel group which comprehensively considering the optimization of ventilation parameters and the dynamic updating of simulation parameters,and the feasibility of the theory and method mentioned above is verified.Taking a complex deep buried diversion tunnel group engineering as the research case,the theories and methods mentioned above are applied to practical project.Firstly,based on the ventilation requirements of complex deep buried diversion tunnel group in different construction periods,different ventilation schemes are formulated;Secondly,based on the fractal analysis of the three-dimensional fractured network,the equivalent thermal conductivity of the surrounding rock is obtained.The simulation methods of ventilation two-phase flow of complex deep buried diversion tunnel group under the consideration of surrounding rock heat transfer is used to simulate the cooling process of ventilation in different ventilation schemes,and then the turbulence structure characteristics of the wind field are analyzed,and the heat transfer of surrounding rock in tunnel,the heat exchange law between surrounding rock and airflow and the regulation of migration and change of pollutants are discussed,Finally,it reveals the internal relationship between ventilation time and excavation length under different construction ventilation schemes,and optimizes the values of simulation ventilation parameters;Thirdly,based on the original time series of site construction parameters,the construction simulation parameters prediction model based on ACDE-SVM is adopted,and the dynamic prediction analysis of the construction simulation parameters of complex deep buried diversion tunnel group is carried out;Lastly,based on the optimized simulation ventilation parameters and the construction simulation parameters prediction time series,the construction simulation optimization model is adopted to realize the construction simulation optimization of complex deep buried diversion tunnel group,and provide theoretical guidance and technical support for the arrangement of construction schedule and the control of site construction period.更多还原