灰色线性回归组合模型在北京地面沉降分层预测中的应用

灰色系统理论GM(1,1)模型,应用于地面沉降模拟和预测中只能分析数据的指数变化规律。对于地面沉降发展过程中,存在的线性关系不能有效地反映。本文利用灰色组合模型中的第一类灰色组合模型即GM(1,1)与线性回归模型相融合。选取北京东部某地面沉降监测站2004～2012年的分层监测数据建立模型,计算出各监测层位沉降的数学模型,并以此预测各监测层位地面沉降量。结果表明：利用灰色线性回归组合模型在对地面沉降进行分层模拟和预测是可行的。在已有数据的基础上,利用数学模型进行沉降模拟时,两种模型的精度均很高,但通过模型预测未来一年沉降量时,灰色线性回归组合模型的精度,要远高于普通均值GM(1,1)模型。     

Prediction of Rock Mass Squeezing of T4 Tunnel in Iran

One of the most important steps in designing underground structures is the evaluation of ground conditions in terms of squeezing potential and behavior of the geological structures. Generally, constructing a tunnel in the squeezing condition is a very slow and difficult task. Therefore, recognition and evaluation of the squeezing potential is very important in selecting a suitable excavation method and support, especially in weak rocks. This research is concerned with the assessment of squeezing potential along tunnel T4 of water conveyance system from Azad dam to Ravansar plain with the length of 11,380 m, located between Kurdistan and Kermanshah provinces, west of Iran. This tunnel is in an almost NS direction and flows through the contact zone of Iran and Arabic plates. According to the engineering geological investigations, the squeezing potential has been recognized as the most important difficulty in the excavation of tunnel T4. This conclusion can be explained by the several indicators including lithology, high disintegration of rock masses, alteration of rocks on the border of Iran and Arabic plates’ contact zone, low rock mass quality, high overburden, and highly jointed rock masses in a shuffle tectonic condition. This paper deals with the engineering geological and geomechanical properties of rock masses. Then, it evaluates squeezing intensity using empirical, semi-empirical, and analytical properties methods. The analysis conducted in these work shows that the tunnel excavation would encounter squeezing problems, which is most severe in region 2 due to the effects of the major young Zagros fault.     

Effects of Rock Bolting on Stress Distribution around Tunnel Using the Elastoplastic Model

INTRODUCTIONThe stability of underground structures duringand after excavation is the most i mportant factor fordesigners because any kind of collapse may destroylarge parts of a finished tunnel,thereby causing ma-jor repairs and ti me loss.To avoid this problem,rock bolts are widely used as ani mportant support el-ement.These are economical,have a short installa-tion ti me,and provide reinforcement to maintain theintegrity of the stressed rock.The effect of the bearing capacity of rock b…     

挤压式盾构隧道施工引起的地表移动及变形

视挤压式盾构施工隧道引起的地表移动为一随机过程,应用随机介质理论对所引起的地表移动和变形进行了分析。工程实例分析表明,该分析方法效果良好。     

Comparison of Squeezing Prediction Methods: A Case Study on Nowsoud Tunnel

According to the International Society of Rock Mechanics, squeezing is a time dependent large deformation occurring during tunnel construction around the tunnel associated with creep caused by exceeding a limiting shear stress (Barla in ISRM News J 2:44–49, 1995). This research is conducted using a case study on the Nowsoud Tunnel, Iran. Being 14 km in length and 4.5 m in diameter, the tunnel is located in the western part of Iran near the Iraq border. Nowsoud tunnel, which was excavated using a double shield TBM, exhibited severe squeezing (with 8919 m) in its critical zone which resulted in excavation termination. In this research, the best approach for predicting squeezing among the recommended methods for reducing the damages caused by squeezing on TBM was determined. In this regard, approaches commonly used to predict squeezing are empirical, semi-empirical, and theoretical–analytical methods. Besides, these methods, numerical modeling is used to estimate convergence generated along the tunnel pathways, which is ultimately used to categorize squeezing. This paper compares squeezing prediction methods in 68 section of Nowsoud Tunnel. These 68 sections indicate that the empirical methods propose a general estimation/overview of squeezing. Among the semi-analytical approaches, the one proposed by Hoek and Marinos (Rock engineering in difficult rock conditions—soft rocks and karst, Taylor & Francis Group, London, pp 49–60, 2000) are compatible with the occurrence of squeezing in the critical zone. However, the degree of predicted squeezing is less than the real degree of squeezing in this zone. Based on the result of Aydan approach, 75 % of the tunnel sections are under squeezing condition. Theoretical–analytical approaches underestimate the possibility of squeezing in the critical zone. Barla?s approach (1995) demonstrated the possibility of squeezing in the critical zone with low intensity. The numerical computations in this paper were performed using Plaxis (version 8.5), a two-dimensional numerical program based on the finite element method. Plaxis results, classified by Hoek and Marinos (2000) method, show that 8800 m of the tunnel length is under the non-squeezing condition. According to all prediction methods, the squeezing zones depend on faulted zones, argillaceous limestone and shale formations such as J₁Kh, J₄Kh, J₅Kh, and Kgr. These formations were identified with a high quantity of shale and argillaceous limestone. Bedding of these geological formations is thin and their geotechnical properties are weaker than those of limestone formations. On the other hand, non-squeezing zones depend on limestone formations such as J₂Kh, J₃Kh, J₆Kh, K^abg, and Kbg. Moreover, all approaches predicted squeezing potential for the critical zone where TBM is jammed.     

电磁驱动可控震源在隧道探测中的应用

地震勘探对地下地质结构的调查是一种行之有效的方法。由于条件的限制,有些地区不适合采用炸药震源,而且炸药震源费用高、危险大。采用电磁驱动的高频可控震源,震源采用全新的电磁驱动方式,频率从10H z到1000H z,向地下发送可变频率振动。为验证高频可控震源野外工作的稳定性、可靠性、有效性,在鞍山东山隧道上进行了实际探测,与已知资料相比,得到了满意的结果。     

The Interaction Between Shield,Ground and Tunnel Support in TBM Tunnelling Through Squeezing Ground

When planning a TBM drive in squeezing ground, the tunnelling engineer faces a complex problem involving a number of conflicting factors. In this respect, numerical analyses represent a helpful decision aid as they provide a quantitative assessment of the effects of key parameters. The present paper investigates the interaction between the shield, ground and tunnel support by means of computational analysis. Emphasis is placed on the boundary condition, which is applied to model the interface between the ground and the shield or tunnel support. The paper also discusses two cases, which illustrate different methodical approaches applied to the assessment of a TBM drive in squeezing ground. The first case history—the Uluabat Tunnel (Turkey)—mainly involves the investigation of TBM design measures aimed at reducing the risk of shield jamming. The second case history—the Faido Section of the Gotthard Base Tunnel (Switzerland)—deals with different types of tunnel support installed behind a gripper TBM.     

隧道施工引起横向地层沉降的随机预测

运用随机介质理论推导的横向地层沉降理论公式计算隧道施工引起的横向地层沉降时 ,影响范围角 β和隧道收敛值ΔR比较难以确定。详细分析了这两个参数的确定问题 ,并通过工程实例验证了横向地层沉降理论公式。     

New Viscoplastic Model for Design Analysis of Tunnels in Squeezing Conditions

Summary  This paper is intended to describe the SHELVIP (Stress Hardening ELastic VIscous Plastic) model, a new viscoplastic constitutive law which has been developed to incorporate the most important features of behaviour observed in tunnels excavated in severe to very severe squeezing conditions. This model couples the elastoplastic and time-dependent behaviour by using a plastic yield surface, as frequently adopted in tunnel design analysis, and the definition of a state of overstress referred to a viscoplastic yield surface. The model is formulated in all its detailed aspects. The related analytical closed-form solution for representing triaxial creep deformations is developed. Also developed is an incremental numerical solution for describing the triaxial stress–strain behaviour under constant strain rate conditions. The model is shown to fit very satisfactorily the results of creep tests on clay shales and relaxation tests on coal specimens, as recently performed for design analysis of tunnels in squeezing conditions. Correspondence: D. Debernardi, Department of Structural and Geotechnical Engineering, Politecnico di Torino, Torino, Italy     

参考作物腾发量预测的径向基函数法

探讨了采用径向基函数网络模型进行参考作物腾发量预测方法的可行性,设计多组数字实验处理研究了输入因子间相关性对网络模型预测准确性的影响,预测结果与Penman-Montieth方法计算结果比较表明,所确定的模型与改进的Penman公式计算值有很高的一致性,具有一定精度。     

隧道系统预测的多因素模型

建立了隧道系统预测的多因素分析模型 ,弥补了以往单因素时序分析的不足 ,并编制了相应的分析程序。结合沈大高速公路韩家岭隧道部分断面的多种实测数据 ,对隧道系统的多因素模型及稳定性进行了研究 ,对相应预测结果进行了讨论     

降雨型浅层滑坡危险性预测模型

通过分析SHALSTAB和TRIGRS等浅层滑坡物理确定性模型存在的问题,提出了基于降雨入渗动态守恒的瞬态降雨入渗模型,该模型考虑了初期降雨过程、降雨历程以及饱和非饱和入渗过程,证明了SHALSTAB模型是该模型的特殊形式,并克服了TRIGRS模型参数繁多及一维入渗路径的问题.将无限边坡模型、瞬态降雨入渗模型和GIS进行耦合,研发了可用于大范围降雨型浅层滑坡危险性预测的集成系统,根据边坡的地质条件、地形参数和降雨特征即可对降雨条件下浅层滑坡的危险性进行评估.     

天津市地面沉降的灰色系统—马尔柯夫预测模型应用

地面沉降速率监测值多呈波形动态,数据波动给预测带来了困难。本文以天津地面沉降多个水准点的监测数据为基本序列,介绍了传统灰色系统模型GM（1,1）的预测过程及结果,并在此基础上进行马尔柯夫处理,建立灰色系统-马尔柯夫预测模型。其结果表明,后者预测精度明显提高,67％的样本误差可降至10％以内,92％的样本预测值误差低于30％。     

利用MDT资料预测油气产能

准确地预测油气产能关系到油气田开发方案的有效制定。目前,常用模块式地层动态测试(MDT)资料进行产能预测。由于MDT的测试时间只有几秒钟到几分钟,因此直接使用MDT测试数据计算储层产量会导致较大的误差。针对这一问题,利用小波分析手段提取了MDT压力测试数据的小波系数之后,采用神经网络方法建立了小波系数与钻杆地层测试(DST)产能间的数学模型,进而用该模型对其产能进行了预测。将所构建的这一整套MDT产能预测方法应用于XJ油田的产能预测中,预测结果与DST测试产能的对比表明,该方法能够较好地对工区内的储层产能进行预测,其预测精度能够满足油气田生产的要求。     

动力荷载作用下软粘土的残余变形计算模式

在试验结果的基础上，提出了动力荷载作用下计算软粘土残余变形的模式。并将计算结果与不排水和部分排水情况下观测到的孔隙水压力及残余应变进行比较，两者具有较好的吻合性。本文所提出的计算模式可用来估算动力荷载（例如地震荷载、交通荷载）作用下建于软粘土地基上的建筑物及其它设施所产生的附加沉降。     

黄河泥沙冲淤数学模型研究

应用作者提出的动床阻力、非均匀沙挟沙力、床沙交换调正及滩槽水沙交换等关系与新的不平衡输沙等初步经验关系,建立了黄河下游及水库河床变形计算一维数学模型。经1960年～1988年共28年实测水沙资料的验证计算,结果与实际基本符合。该模型的特点是可预报各种水流及河床边界条件下的动床阻力;可同时计算确定混合沙中各粒径组的推移质、悬移质挟沙力、含沙浓度垂线分布及上滩水流含沙量;能较好地反映浑水上滩淤积减沙使主槽减淤增冲的实际情况;能够基本反映泥沙多来多排及高含沙水流的冲淤特性;能综合反映出来水流量、含沙量及级配的调节变化产生的输沙冲淤差别。应用该模型计算黄河小浪底水库库区及下游河道的冲淤,成果比较合理。