International Conference on Mechatronics, Control and Electronic Engineering (MCE 2014)

Investigation of Soft Subway Based on Multiple Factors

Li Guihua Zhao Bingshuai School of Earth Science and Engineering School of Earth Science and Engineering Hohai University Hohai University Nanjing, China Nanjing, China [email protected] [email protected]

Tang Wei Guo Xiantao School of Earth Science and Engineering School of Earth Science and Engineering Hohai University Hohai University Nanjing, China Nanjing, China [email protected] [email protected]

Abstract—Objective: the subsidence of soft soil subway key drivers to research the settlement of subway tunnel in tunnel structure is a key threat to the safe operation of soft soil. subway system. In order to detect and respond the problems At present, the subway tunnel settlement monitoring timely, it is necessary to constantly monitor the subsidence of mechanisms during the operation period are not perfect soft soil subway tunnel structure and to analyze its variation enough, with limited capability to meet the demand of over time. Methods: a subway tunnel is buried in the ground practicability, reliability and accuracy of the settlement and is usually in operation all the time, making it difficult to monitoring. Also, it is difficult to meet the long-term use common methods of monitoring the deformation of the demand during the operation of subway tunnel settlement tunnel structure. This difficulty can be overcome by monitoring. As a result, it is urgent to find new research deploying sensors. In this study, based on the characteristics means and methods. This study used a variety of relevant of soft soil subway tunnel subsidence, data of various influencing factors have been monitored. According to the monitoring information as fundamental factors, analyzed principal component analysis and the qualitative analysis of the degree of influence of the factors, and studied the different factors, a mathematical relationship was derived subway tunnel settlement cause during the operation between each influencing factor and soft soil subway tunnel period. subsidence. By applying principal component analysis from both time and space, the degree of influence of each factor II. ANALYSIS OF THE CAUSE OF SOFT SOIL SUBWAY was analyzed both qualitatively and quantitatively. Results: TUNNEL SUBSIDENCE the results accurately showed the influence of different According to the structural characteristics and settling factors in the process of subway tunnel subsidence. characteristics of subway tunnel in soft soil, the factors Conclusion: this study shows that the use of sensors could be influencing the settlement of subway tunnel structure in a practical and effective method to monitor the settlement of service period are divided into four categories as, soft soil subway tunnel. environment factors of soil, tunnel structure factors, train Keywords- subway tunnel; subsidence; multiple factors; load factors and factors of external environment load [4-7]. soft soil; principal component analysis The environment factors of soil includes geological factors and factors about the variation in level; the I. INTRODUCTION tunnel structure factor is mainly the leakage in caused by cracking; the train load factors include train The adverse conditions of the subway tunnel in soft traffic and passenger volume; and the factors of external soil, such as stratum geological environment, complex environment load mainly consist of building load, traffic engineering characteristics and significant spatial variation load, regional land subsidence, and regional atmospheric of soil distribution, have posed a serious threat to the precipitation on the ground area above the tunnel. settlement of tunnel structure [1-3]. At the same time, some other factors, such as the external environmental A. Soil Environmental Factor influence, the change of the load and other unforeseen Soft soil is characterized by low strength, variable causes, have increased the complexity of the tunnel thickness and complex characteristics of consolidation and structure settlement. Therefore, identification of the cause creep. Also, a large difference exists between the vertical of the subway tunnel settlement in the soft soil, and and the horizontal permeability. The soil environmental analysis of the qualitative and quantitative extent of the factors can be grouped into two categories, namely, settlement caused by different influencing factors are the geological factors and the underground water level change. The geological factors can be further divided as, the tunnel

© 2014. The authors - Published by Atlantis Press 741

under the bottom of the soil distribution, and the aging st =+Ctln d (3) ( and creep) characteristics. On the other 2sC In equation (3), v（constant for a cross- hand, the underground water level factor represents the soil C=2 subsidence caused by the variation in groundwater level. He section). Equation(3) proves that there is a linear 1) Geological Factor relationship between the amount of consolidation a) Effect of the substratum soil layer distribution on settlement and ln t at time t. the baseboard of a tunnel 2) Variation in groundwater level When the substratum soil layer distribution on the According to Terzaghi’s effective principle, soil baseboard of a tunnel is thick, the tunnel experiences a compression after lowering the groundwater level to any large settlement, and vice versa. Therefore, the thickness depth Hcan be obtained as: of soft soil layer directly affects the settlement stability of the longitudinal subway structure. Based on the layer-wise HHHDe aD-pa()yhg summation method, the settlement for S===dyvdyvwdy (4) òòòhhh unidirectional compression is given by the formula[8]： 1+e011++ee00

nnDp where eis the initial of soil; Deis the S=DSH=×i (1) 0 ååiiE ii==11Si change in the void ratio; avis the compressibility of soil; g is the pore water ; h is the initial where is the number of thin compressed soil layers w i groundwater level; and His the lowered groundwater from 1 to n; His the thickness of the layer; Dpis the level.

average additional stress on the layer; and ESis the For a point on the soft soil substratum on the floor of compression modulus of the soil layer. the subway tunnel, each parameter in equation (4) can be For the case of a thin compressed soil layer, where the considered as constant, except H. Hence, by using the additional stress and the average compression modulus are integral method, we get: fairly constant, equation (1) shows that the foundation settlement is proportional to the thickness of soil. Test data 2 S=a1H++a23Ha (5) showed that the additional stress and the soil compression modulus at the tunnel substratum at a cross-section agaggha remained the same before and after the construction of a vwvwvw 2 where a1=,,a23=-=ah. tunnel in soft soil. Even for different nearby sections, the 2(1+e0)1++ee002(1) change in additional stress of tunnel substratum soil is Equation (5) shows that soil subsidence is a second insignificant because the change in the longitudinal degree polynomial of groundwater level. of the tunnel is small. However, the soil compression modulus may exhibit certain differences because of the B. Load Factors of Train Operation geological inhomogeneity along the length of the tunnel. Under the action of train-induced dynamic load, soft Thus, the amount of compression of substratum on soil has elastic deformation and plastic deformation. The different fracture surfaces follows a linear relationship former does not affect the post-construction settlement but

between S and HE/Si . the latter contributes to the tunnel settlement. By a preliminary study, the relationship between the soft soil b) The effect of aging (soil consolidation and creep) [9] deformation and the train load is derived as: Deformation of soft soil is actually a process of mn consolidation and creep simultaneously. According to the æqqöæö E=+aNçds÷ç÷1 b (6) soil unidirectional consolidation theory： pç÷ç÷ èqqfføèø p2 ì æö- T 8 4 v ïs=se×-ç÷1 where Epis cumulative plastic deformation. ït ç÷p2 (2) íèø N,qd,qf,and qsare respectively the number of cyclic ïCtv ïTv = îH2 loading, the dynamic stress of traffic load（ kPa ）, the （ ） In equation (2), final settlement of foundation S is a static destructive stress kPa , and the static deviator theoretically determined value which is independent of stress. Parameters a,b,,mn are constants. time; C is the coefficient of consolidation and is a Equation (6) shows that soil parameters, static deviator v stress and static destructive stress are constant at the same constant; and His the thickness of compressible stratum location. For subway tunnel, the passenger traffic and the and is a constant. The consolidation settlement of the traffic frequency are relatively stable on the section for foundation, St, is a function of time (t). However, the final non-transportation hub. Thus, the dynamic stress of traffic settlement needs to be a theoretical value which is difficult load in certain cases can also be regarded as a constant. to measure in reality. Also, huge complexity and Consequently, equation (6) can be adapted as follows: uncertainty are involved in acquiring the value of mb consolidation coefficient and the thickness of compressible Epd=ABqN (7) stratum. Thus, by simplifying and deducing equation (2), we can obtain:

742

n long-term basic constant, which is a vertical constant and 1 æöq where A=,1Ba=+ç÷s are constants. the other factors of each are different. m ç÷ qqffèø

C. Effect of Tunnel Leakage TABLE I. THE CONTRAST TABLE OF VARIATION CHARACTERISTICS Apart from the direct erosion of the loose soil variation surrounding a tunnel, leakage can create soil sediment flow No. Factor Lateral Longitudinal characteristics paths, reduce , and increase soil Substratum . Consequently, soil gets compacted to soil layer per issue distribution longitudinal cause subsidence, which in turn leads to the settlement of 1 the variable on the variation the subway tunnel. Hence, leakage is a serious threat to the same safe operation of a tunnel structure. baseboard of a tunnel According to the consolidation theory, The amount of Soil per issue lateral soil consolidation, S, is calculated by storage rate, S, and 2 Consolidation the variable t s variation the effective stress. The formula is as follows: and Creep difference bidirectional Ss per issue variation S=mDssH=DH=DShH (8) Groundwater Each point tvzzs 3 the and dominant g Level the difference w difference lateral where Dhis the leakage caused by variation in water variation per issue Each point head and His the soil thickness. Since soil water storage lateral 4 Train Load the the same of a and soil thickness are constant for a location, equation (8) variation shows that soil compression is directly proportional to difference section per issue bidirectional Tunnel Each point water head. 5 the variation leakage the difference D. The influence of the external environmental factor difference bidirectional Regional per issue Each point variation The external environmental factors mainly include the 6 Building the the difference and change construction load around the subway tunnel area, the Load difference traffic load of the ground, the regional land subsidence, shortly bidirectional Regional per issue and the regional atmospheric precipitation. Among all Each point variation 7 Land the those factors, two main factors are the construction load the difference and lateral Subsidence difference around the subway tunnel area and the regional land change greatly subsidence. The other factors are minor and indirect; they OF SUBSIDENCE CAUSES have little effect on the subway tunnel settlement and can be ignored. IV. ANALYSIS OF INFLUENCED DEGREE OF SUBWAY 1) The construction load around the subway tunnel TUNNEL SETTLEMENT area In terms of the influence in the settlement process and The effect of the variation in construction load around its nature, the influencing factors are so many that there the subway tunnel area to the settlement of a point is must be some correlation with each other and overlap dependent only on the load. Thus, the settlement maintains between the information they carry. Thus, the principal a linear relationship with the vertical stress as: component analysis is applied in this study [10-11]. Given s=aF (9) the variation of settlement factors, the settlement factors where s is the point of settlement, a is a coefficient, and can be discussed and analyzed from two aspects as, F is the construction load. horizontal and vertical. The analysis can also be performed 2) The regional land subsidence in the same way for the measurement at the same point at A deep multilayer soil generates compressed different periods, and measurement at different points at deformation under the influence of various factors, which the same period. cause regional land subsidence. Typically, a subway tunnel is in shallow ground, and undergoes settlement A. Horizontal Analysis corresponding to the ground subsidence. Therefore, the Under the circumstance that the thickness of soil settlement of subway tunnel structure and regional land substratum on the floor of the subway remains essentially subsidence follow a linear relationship. constant, the same point was considered at different measurement periods. The change in each factor at III. VARIATION CHARACTERISTICS OF THE INFLUENCE different measurement period was analyzed to estimate the FACTORS IN SETTLEMENT OF SUBWAY TUNNEL IN SOFT SOIL influence of individual factor to the degree of settlement. TableⅠpresents the influences of the different factors Thus, the analysis included monitoring data from 5 discussed in section II to the long term settlement of a measuring points of a subway tunnel for 11 consecutive subway tunnel in the lateral and longitudinal directions. periods (over 3 years). Since the thickness of soil As Table 1 shows, for the same period of monitoring, substratum cannot be included into principal component all point (i.e. soil consolidation and creep) agreement. The analysis at each point, this study took the aging, the square other factors are not identical in addition to the same of groundwater level, groundwater level, train load, and section of train operation load, For the same monitoring cumulative ground subsidence into consideration. The point, the thickness of tunnel floor bottom layer remains contributions of different factors at different points are listed in TableⅡ.

743

the time, the impact of various factors on the tunnel gradually dispersed. On the one hand, the influence of the main factors, such as the thickness of soil substratum and Thickness Trai Cumulative of Soil Groundwater the groundwater level, remains unchanged. On the other n Ground Measuremen Substratu Level hand, the influence of the rest of the factors gradually Load Subsidence t Period m increases. 2 b* HE/ H 2 H H 1 S 2 N 4 V. CONCLUSIONS 34.78 2007.3 59.998 4.925 0.288 2.34E-015 9 This study identified seven different factors that 33.71 influence the settlement of subway tunnel. A mathematical 2007.9 61.798 4.213 0.273 1.49E-015 6 model was developed based on the principal component 33.24 2008.3 62.290 4.145 0.320 1.03E-015 analysis method to investigate the degree of influence of 5 the factors both qualitatively and quantitatively. The 33.46 2008.9 61.947 4.228 0.356 1.26E-015 analysis results were interpreted from the two aspects, 8 33.63 namely, horizontal (time) and vertical (space). The 2009.3 61.645 4.306 0.419 3.42E-015 0 following conclusions are made: 32.93 · In overall, tunnel floor bottom layer thickness and 2009.9 62.415 4.219 0.428 4.32E-015 8 soil consolidation and creep are the main factors to influence the settlement of a subway tunnel. TABLE II. THE DEGREE OF CONTRIBUTION OF DIFFERENT FACTORS Second most influence was occurred due to AT THE TYPICAL MEASUREMENT POINTS（UNIT：%） underground water level. The influence of train It can be concluded from Table 2 that the aging (soil loads and the accumulated settlement is relatively consolidation and creep) is the most influencing factor; the small. effect of groundwater level is relatively small and the · For a single measurement point, aging (soil remaining factors have negligible effect. consolidation and creep) is the main influencing factor; the influence of underground water level is B. Vertical analysis fairly small, and the influences of train loads and Under the circumstance that the aging is essentially the accumulated settlement are negligible. constant to all the points at a particular measurement · For a single measurement period, tunnel floor period, analysis was performed for the other factors to bottom layer thickness is the main influencing estimate the degree of settlement. Since the aging cannot factor; the second influencing factor is the be included into the principal component analysis, this underground water level; the third and the fourth study took the thickness of soil substratum, the square of are respectively the train loads and the groundwater level, groundwater level, train load and accumulated settlement. cumulative ground subsidence into consideration. The · With an increase in service life, the influence of contributions of different factors at different periods are the thickness of tunnel floor bottom layer shows a listed in TableⅢ. weak fluctuating trend. The position of its main factors have an increasing trend, the influence of TABLE III. THE DEGREE OF CONTRIBUTION OF DIFFERENT FACTORS underground water level has a similar change, but AT DIFFERENT MEASUREMENT PERIODS（UNIT：%） changes of train loads and the accumulated Cumulative Groundwater Train settlement are more and more obvious. Aging Ground Measuring Level Load · The principal component analysis method for Subsidence point analysis of the causes of settlement in service life * ln t H 2 H b H of subway tunnel in soft soil has a good 2 2 N 4 applicability. It can meet the requirements of K1+033 83.154 16.758 0.069 0.016 0.003 qualitative and quantitative analysis of the degree K1+090 83.936 15.961 0.073 0.028 0.003 of influence of various factors. The method is able to qualitatively and quantitatively analyze the K1+337 82.576 17.345 0.060 0.017 0.002 degree of influence of different factors from the two aspects, namely, horizontal and vertical. K1+470 74.187 25.600 0.172 0.033 0.008 ACKNOWLEDGMENT K1+641 62.178 37.588 0.180 0.032 0.022 This project was supported by Hohai University TableⅢ shows that the thickness of soil substratum fundamental research funds for the Central Universities remains the main factor to affect the settlement in the project(2012B01714). transverse direction. The effect of groundwater level is the second most influencing factor. The effect of train load on REFERENCES the settlement is relatively small and that of cumulative [1] Jiwei Chen,Longxi Zhan. Regularity analysis and deformation ground subsidence is negligible. measurement of Shanghai Metro Line 1[J]. Urban In the longitudinal direction too, the thickness of soil ,2000,(2):51-56. substratum on the floor of the subway is the most [2] Rulu Wang,Xianhao Zhou,Yongliang Yu. Problems and Solutions influencing factor, and the remaining factors have similar of Shanghai Metro Monitoring recently[A]. The Fourteenth orders of influence as with the transverse direction. Over Symposium of Professional Committee of the underground railroad,

744

Institute of tunnel and underground engineering, China Civil valley flat [J]. Engineering Geology and for civil Engineering Society [C].2001:239-242. engineering majors,2006,(3):112-116. [3] Guanlong Huang,Ming Wei,Aiming Han,Guoxiong Mei. Analysis [7] MingLiu,Maosong Huang,Jinjun Li.Long-term Settlement of on the subsidence of tunnel foundation in Nanjing Yangtze River Saturated Soft under Subway Loading[J].Chinese Journal of valley flat [J]. Engineering Geology and Hydrogeology for civil Underground Space and Engineering, 2006, 2(5): 813-817. engineering majors,2006,(3):112-116. [8] Mengxi Zhang.Principle of [M].Wuhan: Press of [4] Yongguo Lin,Shaoming Liao,Guobing Liu. A Discussion of the Huazhong University of Science and Technology,2007. Factors Effecting on Longitudinal Deformation of Subway [9] Cha,i J. and Miura, N. Traffic-load-induced permanent deformation Tunnel[J]. Underground Space,2000,20(4):264-290. of on soft [J]. Journal of geotechnical and [5] Rulu Wang. Factors Influencing Deformation of Shanghai Soft Soil geoenvironmental, 2002: 907-915. MetroTunnel and Deformation Analysis [J]. Substructure Work [10] Xiaoqun He. Multivariate statistical analysis (The second edition) and Tunnel,2009,(1):1-6. [M].Beijing: Press of Renmin University of China, 2008. [6] Guanlong Huang,Ming Wei,Aiming Han,Guoxiong Mei. Analysis [11] Sun Yifei,Liu Hanlong, Yang Gui,Xiao Yang. Formulation of on the subsidence of tunnel foundation in Nanjing Yangtze River cross-anisotropic failure criterion for [J].Water Science and Engineering, 2013,(6):456-468.

745