- Email: [email protected]
Stress and deformation characteristics of transmission tower foundations in permafrost regions along the Qinghai–Tibet Power Transmission Line
Stress and deformation characteristics of transmission tower foundations in permafrost regions along the Qinghai-Tibet Power Transmission Line Zhi Wen, Qihao Yu, Mingli Zhang, Ke Xue, Liangzhi Chen, Desheng Li PII: DOI: Reference:
S0165-232X(15)00123-8 doi: 10.1016/j.coldregions.2015.06.007 COLTEC 2118
To appear in:
Cold Regions Science and Technology
Received date: Revised date: Accepted date:
28 September 2014 2 June 2015 9 June 2015
Please cite this article as: Wen, Zhi, Yu, Qihao, Zhang, Mingli, Xue, Ke, Chen, Liangzhi, Li, Desheng, Stress and deformation characteristics of transmission tower foundations in permafrost regions along the Qinghai-Tibet Power Transmission Line, Cold Regions Science and Technology (2015), doi: 10.1016/j.coldregions.2015.06.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Stress and deformation characteristics of transmission tower foundations in permafrost
*
, QihaoYu, Mingli Zhang, Ke Xue, Liangzhi Chen, Desheng Li
SC R
IP
Zhi Wen
T
regions along the Qinghai-Tibet Power Transmission Line
(State Key Laboratory of Frozen Soil Engineering, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences,
NU
Lanzhou Gansu 730000, China)
MA
ABSTRACT: A transmission tower foundation embedded in frozen soil is subject to both the wind-induced uplift and frost heave forces. The frost heave results in an upward force acting on
TE
D
the foundation, while additional stress induced by the structure load may compress the underlying soils. The freezing- and thawing-induced deformations tend to cause further structural loads and
CE P
lead to instability problems within the structure. To evaluate the engineering risk and ensure the safety of the Qinghai-Tibet Power Transmission Line (QTPTL) system, stress sensors were
AC
installed at the bases of two test tower foundations to investigate the stress state of the tower foundations. Using data on air and ground temperatures, and the deformation of tower foundations, we analyzed the stress variation, and the causes were discussed here. The results showed that the stresses at the bases of tower foundations had a close relationship with air and ground temperatures. The cooling of the underlying soils led to the occurrence of frost heave, which pushed the foundations upward and caused a significant stress bulb under the bases of tower foundations. Seasonal variations in the contact stress depended on the seasonal freezing and
*
Corresponding author: Tel.: +0086-931-496-7299. E-mail address: [email protected](Z. Wen) 1
ACCEPTED MANUSCRIPT thawing of foundation soil. The contact stress increased with the cooling of the underlying soils and decreased with the warming of the underlying soils. The results also showed that the contact
IP
T
stress was free of the wind influence, i.e., the wind-induced uplift force was minor for the contact
SC R
stress. To fully understand the influences of freezing and thawing on the stress state of tower foundations, a thermal-elastico-plastic finite element model for the tower foundation-soil system was established, and the stresses and deformations of a tower foundation subject to frost heave and
NU
thaw settlement were simulated. The results showed that the frost heave force induced by soil
MA
freezing potentially threatens the safety and normal operation of the QTPTL. Thaw settlement may lead to harmful deformation of tower foundations if global warming is considered in the
D
numerical model. The remedial measure with thermosyphons only can reduce the settlement of the
TE
foundation and will increase the frost jacking risk of the foundation.
CE P
KEY WORDS: Qinghai-Tibetan Plateau, stress, tower foundation, frost heave force, transmission
AC
line, permafrost.
1 Introduction
The Qinghai-Tibet Power Transmission Line (QTPTL) runs across 1,038 km of permafrost and seasonally frozen ground in the Interior of the Qinghai-Tibetan Plateau. The mean annual air temperature of permafrost and seasonally frozen ground along the QTPTL varies between -3C and -7C and the minimum air temperature is lower than -37C in short durations. The active layer is subjected to annual freeze-thaw cycles and its thickness varies between 2 and 3 m. Thus, substantial heave force is expected due to the existence of extensive frost-susceptible soils and cold climate. Moreover, the wind-induced uplift force is another load for the transmission towers
2
ACCEPTED MANUSCRIPT and can reach more than 1,500 kN. Warm and ice-rich permafrost is widespread in permafrost regions of the Qinghai-Tibetan Plateau. The warming and subsequent thawing of ice-rich
IP
T
permafrost tends to result in serious engineering and environmental consequences. If not well
SC R
designed or built, the tower foundations may be jacked up in the freezing periods and subside in the thawing periods, which would result in expensive maintenance costs and significantly threatens the safety and operation of the transmission lines. In arctic and sub-arctic regions,
NU
intense frost-induced heaving has considerably deteriorated the condition of numerous power
MA
network objects and significantly threatens the construction and operation of transmission lines (e.g., Lyazgin et al. 2004; Jiang and Liu 2006; Cheng et al. 2009). Seasonal frost heave or thaw
D
settlement deformation can impact the mechanical state of the foundations, while the stress state
TE
of tower foundations and surrounding soils has a close relationship with foundation failures.
damage.
CE P
Therefore, it is important to understand how the stress changes in order to reduce tower foundation
AC
Previous studies of foundations in cold regions have been focused on the adfreeze bond strength and frost heave force (e.g., Penner, 1974; Tong and Guan, 1985; Johnson and Esch, 1985). Since the 1930s, measurements of the adfreeze strength between piles and frozen ground have been undertaken (Saitykov, 1944; Trow, 1955; Penner and Irwin, 1969; Penner and Gold, 1971; Sadovsikiy, 1973; Perameswaran, 1978; Ladanyi and Foriero, 1998). Many studies focused on the bearing capacity and uplift performance of piles in frozen soil in recent decades (Selvadurai and Hu, 1996; Wang R et al., 2005; Wang X et al., 2005; Aksenov and Kistanov, 2008; Zhang et al., 2008). To further understand the frozen soil-foundation interactions, researchers have used the finite element method and model testing to determine the distributive features of stress and 3
ACCEPTED MANUSCRIPT deformation. Wu et al. (2010) proposed a nonlinear elastico-plastic finite element model for a pipeline-soil system and calculated the mechanical performance of the oil pipeline subject to
IP
T
differential frost heave in permafrost regions. Xu et al. (2010) investigated the stress and strain of
SC R
a buried pipeline and surrounding soils by a model test, and analyzed the ambient temperature of the pipeline foundation, the displacement and axial strain, and the stress in the pipeline induced by frost heave and thaw settlement. However, the stress and deformation characteristics of tower
NU
foundations have received little attention and there has been limited effort towards the
MA
understanding of the mechanical behavior of foundations.
D
To investigate the stresses and deformation of tower foundations in the permafrost regions along
TE
the QTPTL, the thermal regime, stress state, and deformation were monitored and simulated. The results were analyzed to understand the frost damage processes of tower foundations, as well as to
AC
in cold regions.
CE P
make mitigative recommendations for the design and the construction of power transmission lines
2 On-site experimental investigations on stress and deformation of tower foundations
2.1 Site description and instrumentation
To study the stress and deformation of tower foundations in cold regions, two experimental tower foundations were constructed at Qingshui’he along the QTPTL in 2011. The elevation of the study site is 4,465 m a.s.l. and the average annual air temperature during the period of 2011 to 2012 at the site is -4.9 C, with extremes of approximately 19.97 and -38.04 C. Weather records show that the average annual wind speed is 4.8 m/s. Engineering geological
4
ACCEPTED MANUSCRIPT investigations at the site indicate that there is ice-rich clayey soil with a thickness of 0.5 to 4 m below the permafrost table and the ice content by volume is about 50%~80%. The mean
IP
T
annual ground temperature (MAGT) of the permafrost at the 490# test tower foundation is
SC R
-0.5 C and the permafrost table is 2.5 m. The MAGT of the permafrost at the 492# test tower foundation is -1.0 C and the permafrost table is 2.8 m. The thickness of ice-rich permafrost at 490# and 492# is 2.2 m and 3.0 m, respectively. The test tower foundations are located in a
NU
warm and ice-rich permafrost zone and have low thermal stability.
MA
The contact stresses at the soil-foundation interface are vital to understand the interaction between
D
structural foundations and the supporting soil. To investigate the contact stress characteristics of
TE
tower foundations, three thin-film pressure sensors were installed beneath each tower foundation. To monitor ground temperatures at the test foundation section, six boreholes were drilled in the
CE P
vicinity of the tower foundation soils (Fig. 1). The depths of these boreholes varied from 10 to 23 m. In each borehole, thermistors were installed at depth intervals of 0.5 m from the ground surface.
AC
The thermistors were made by the State Key Laboratory of Frozen Soil Engineering, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, and with a calibrated accuracy of ±0.05 C. The deformation of the tower foundations was measured by a water level instrument.
To study the cooling effect of two-phase closed thermosyphons, four thermosyphons were installed around the 492# tower foundation. The length of thermosyphons was 9 m; the evaporation part was 7 m-long and was embedded in the ground. The condensation part was 2 m-long and was exposed in the air. Fig. 1 shows the sensor installation design at the test footing
5
ACCEPTED MANUSCRIPT foundation and force analysis.
IP
T
3 Results and discussions
SC R
3.1 Seasonal variation of contact stress
Fig. 2a shows the seasonal variation of contact stress under the 490# test tower foundation. The
NU
contact stresses of 490# showed significant variations. The absolute value of contact stress decreased gradually after October and reached approximately 0 kPa in January. Then the contact
MA
stress increased gradually to approximately 300 kPa in May, and then it remained constant during the summer. The seasonal variation of contact stress had a close relationship with the seasonal
TE
D
variation in air temperature near the site (Fig. 2c). The contact stress decreased significantly with the decrease in air temperature in winter and increased with the increasing air temperature. The
CE P
contact stress showed a positive correlation with the seasonal variations of ground temperatures near the tower foundation (Fig. 3a). The tower foundation tended to suffer from the force induced
AC
by wind. However, our experimental results showed that the seasonal variations contact stress beneath the foundation did not appear to have a visible relationship with the wind speed near the site (Fig. 2d).
The 490# tower foundation is located in an ice-rich permafrost site with a MAGT of -0.5 C. The ground temperatures near the base of tower foundation show significant seasonal variation. The compression of warm permafrost leads to the occurrence of settlement deformation. Our monitoring showed that settlement deformation at 490# has occurred since 2010, and total settlement amount reached 0.08 m in the past three years (Fig. 3b). As shown in Fig. 1(b), the
6
ACCEPTED MANUSCRIPT contact stress depends on the load, frost heaving force, and adfreeze bond strength between the foundation surface and frozen ground. Sub-surface soil freezing causes the variation of tangential
IP
T
frost heaving force and adfreeze bond strength between the foundation surface and frozen ground.
SC R
The increase in tangential frost heaving force and adfreeze bond strength during soil freezing may offset the contact stress, which results in the decrease in contact stress in cold season. During thawing season, the disappearance of tangential frost heaving force and the decrease in adfreeze
NU
bond caused the increase in contact stress. Thus, the contact stress at 490# depends on the air
MA
temperature or ground temperature. Our monitoring showed that contact stress at 490# reached approximately 0 kPa in January, which implies that the tower foundation has a potential safety
D
hazard to be jacked out of the ground. The deformation observation showed that uplift does not
TE
occur due to the existence of adfreeze bond. To ensure the safety of the tower foundation, it is
measures.
CE P
necessary to continue the stress and deformation observations or to apply some anti-frost action
AC
Fig. 2(b) shows the seasonal variations in contact stress under the 492# test tower foundation. The value of contact stress at 492# is much higher than that at 490#. However, the seasonal variability of the contact stress at the tower 492# was much smaller than that at 490#. The contact stress increased linearly since May 2011 due to line hanging, and reached its peak in August 2011, after which it remained largely constant. In contrast to that at the tower 490#, the contact stress at 492# did not display significant seasonal variations and was free from the influence of air temperature and wind speed. Due to the cooling effect of thermosyphons, ground temperatures near the tower foundation clearly dropped since May 2011.The ground temperatures decreased from -0.5C in May 2011 to -2C in May 2013 (Fig. 3a). The monitoring of tower foundation deformation 7
ACCEPTED MANUSCRIPT showed that a slight frost heave occurred at 492#, and reached 0.02 m in the past three years (Fig.
T
3b).
IP
The 492# tower foundation is located in an ice-rich permafrost site with a MAGT of -1.0 C. The
SC R
abundant water at the site and the cooling effect of thermosyphons led to the occurrence of normal frost heaving force, which pushed the tower foundation upward. The resistance of load and
NU
adfreeze strength between the foundation surface and frozen ground limits the frost heave and
MA
causes high contact stress at 492# test foundation.
It is well known that contact stress depends on the load, wind-induced force, and frost heaving
D
force. The static load of the upper structure does not vary significantly with time. The
TE
wind-induced force includes uplift force and down force, depending on wind direction. In this
CE P
study area the wind direction and wind speed vary quickly during the day, but the annual variation of wind speed is not significant. Therefore, the wind did not have an obvious influence on the
AC
seasonal variations of the contact stress. The influence of frost heaving force on a tower foundation includes two parts: the tangential frost heaving force induced by soil freezing of the active layer, which partially offset the contact stress, and the normal frost heaving force induced by soil freezing under the base, which augments the contact stress.
3.2 Daily variations of contact stress
The contact stresses at 490# and 492# both showed significant daily variation. The relationships of the contact stress with air temperatures, ground temperatures, and wind speed were analyzed, respectively. Fig. 4(a) shows the daily variations in wind speed and air temperature near the site.
8
ACCEPTED MANUSCRIPT Observational data indicate that the variations in wind speed had a similar trend with that of air temperature in most instances. Wind speed increased gradually with the increase in air temperature
IP
T
during the day, whereas it decreased gradually with the drop of air temperature during the night. If
SC R
there is no disturbance of large-scale circulation, both wind speed and air temperature depend on the variation of solar radiation. Thus, in this study the daily variability in wind speed was similar
NU
to those in air temperatures in most cases.
Fig. 4(b) shows the daily variations in wind speed and in the contact stress at 490# during January
MA
2012. The results show that the wind speed in the first half of January 2012 was low but was
D
fluctuatingly greater in the second half. However, the daily variations of the contact stress were
TE
similar during the entire January, suggesting that contact stress was free from the influence of
stress.
CE P
wind speed. We therefore conclude that the wind speed had no significant impact on the contact
AC
Fig. 4(c) shows the daily variations in air temperature and in the contact stress at 490#. The results show that the daily variations in air temperatures had a close relationship with those of the contact stress. The contact stress responded quickly to the variations in air temperatures (Fig. 4(d)). The observational data at 492# also showed a similar rule (Fig. 5).It was deduced that the impact of air temperature on the contact stress may be related to the variations in tensile force caused by the thermal expansion and contraction of the transmission line system.
3.3 Numerical simulation analysis on thermal-mechanical process of tower foundations in permafrost regions
9
ACCEPTED MANUSCRIPT 3.3.1 Computation model and parameters
T
A thermal elasto-plastic computation model by Wen et al. (2010) was adopted to further study the
IP
thermal regime, stress variations and deformation characteristics of tower foundations in
SC R
permafrost regions.
3.3.1.1 Thermal regime computation model
NU
If the water infiltration into, and the water movement in the active layer are ignored, the heat
f
T T 1 T ( ) ( r ) t z z r r r
T f n
u
Tu Lw0 n t
T
c
the media;
(2）
u indicate
(3）
the frozen and thawed states of soils, respectively.
—specific heat capacity of the soil;
—temperature;
AC
conductivity;
and
f
CE P
—soil density;
TE
T f ( (t ), t ) Tu ( (t ), t ) Tm
where the subscripts
(1）
D
c
MA
conduction with phase change in frozen ground can be described as follows:
L —latent
r
and
z
—time,
—coefficient of heat
heat；w0 —the initial water content；Tm —frozen point of
— the frozen and thawed interface;
thawed interface; and,
t
n—
the direction vector of the frozen and
are the radial and vertical distance in cylindrical coordinates,
respectively. The conditions at the fixed boundaries are: T Ta
Where
Ta is
T a(Ta T ) n
soil temperature and
a is
(4）
a coefficient of heat convection between thermosyphon and
air. 3.3.1.2 Mechanics computation model 10
ACCEPTED MANUSCRIPT When the influence of soil temperature on soil deformation is considered, and the influence of soil
can be expressed as follows：
IP
Total strain
T
deformation on soil temperature is ignored, the governing equation can be expressed as follows
e p th ry f
SC R
(5）
Bu
strain;
B —the transformed matrix； —elasticity strain; e
T
ry —the
(7）
p —plastic
compression strain of thawing soil or frost heave strain;
MA
Where
NU
D
(6）
strain;
f
th —thermal
—the compression
strain of frozen soil; and D —total material factors matrix.
TE
D
1) Elastic strain
The elastic deformation obeys the general Hooke’s law:
D e
e 1
(8）
CE P
De e
(9）
AC
where： —stress matrix; e —strain matrix;
D —elasticity. e
For the plane strain problem, 1 E ( 1 ) [De ] (1 )(1 2 ) 1 0
1 1 0
0 1 2 2(1 ) 0
(10）
where E is elastic modulus, is Poisson’s ratio, and they both are functions of temperature T. 2) Plastic strain Plastic strain increment relates to not only to the stress increment, but also the stress state. The
plastic strain increment { ij p } can be expressed as follows： 11
ACCEPTED MANUSCRIPT
{ ij } p
Q ij
（11）
Adopting the associated flow law, [ D p ] can be expressed as follows：
T
S S 2 S3
S 32
（12）
Q Q y 1 x
S3
1 2 Q 2(1 ) xy
S0 S1
Q Q Q S2 S3 x y xy
（13） （14） （15） (16）
MA
S2
SC R
Q Q x 1 y
NU
S1
2 2
IP
S12 1 [D ] S1S 2 So S1S 3 p
Thus, the elasto-plastic matrix [ Dep ] can be expressed as follows：
D D D e
p
(17）
D
ep
CE P
3) Thermal strain
TE
where is flow parameter；Q is plastic function, and is shear stress.
A change in soil temperature will result in a soil volume change, and the expansion ratio or
AC
shrinkage ratio th can be expressed as follows:
th T
where
is linear expand coefficient ( 1 / C );
（18） T is current temperature (℃).
4) Thaw settlement and compression strain Frozen ground is a 4-component and 3-phase complex material composed of liquid water, ice, gas, and soil grains. It has special components, such as ice crystals, lens and ice interlayers. The ice will melt if the temperature exceeds the freezing point of soils, the thawing soil will consolidate and the soil grains will realign due to self-weight. The above mentioned properties of frozen soils are generally called thaw settlement and the deformation is generally called 12
ACCEPTED MANUSCRIPT thaw-settlement deformation. Thawing soils will be further compressed if a load is applied. The thaw settlement and compression of the thawed soils are generally expressed by thaw
IP
T
settlement and compression coefficients, which are related to the water content, dry density,
SC R
and soil type. The thaw settlement and compression strain ry can be expressed as follows: ry Ar Ay P
（19）
where Ar , Ay and P are thaw settlement coefficients (%), compression coefficient (MPa-1) and
NU
load (MPa), respectively.
MA
5) Frost heave strain
The impact of load on frost heave was ignored in the model due to small load. So the frost heave
D
is expressed by frost heave coefficients related to soil type and water supply condition. The frost
（20）
CE P
ry Br
TE
heave strain ry can be expressed as follows:
AC
where Br is frost heave coefficients (%).
3.3.2 Computation parameters
Fig. 6 shows the geometry of a typical axis-symmetrical footing foundation along the QTPTL. An active layer thickness of 2.5 m is assumed for a typical site. The site has ice-rich permafrost with a thickness of 5.4 m below the permafrost table and the moisture content is 50%. The ice-poor permafrost is located at a depth of 7.9 to 20.0 m. The thermal parameters, initial temperature conditions, and the finite element method used for simulating the thermal regimes were similar to those in Wen et al. (2005).The MAGT of the upper boundary condition was 1.0 °C, and the annual range of the ground surface temperature was 26 °C. The annual mean concrete surface 13
ACCEPTED MANUSCRIPT temperature of the upper boundary condition was set at 1.5 °C, and the annual range was 30 °C. The surface temperature varied in a modified attenuated sine wave. The lower boundary had a heat
IP
T
flux constant of 0.06 W·m−2. The vertical axis of the foundation had symmetrical boundary
SC R
condition, and the other side boundaries were adiabatic. Table 1 lists the thermal properties of the materials in the computational domain, in which the symbols “” and “+” denote frozen and the
NU
unfrozen states, respectively.
The thermal exchange of the thermosyphon is assumed to be:
MA
Q F T (t )
(21)
where Q is the heat exchange, T (t ) the temperature difference between environment and the
D
the coefficient of convection heat transfer, and F the effective area of the
TE
soil around,
radiator .The thermal resistance of the thermosyphon was not taken into account in computation
CE P
because it has a thermal conductivity capacity several orders higher than that of the soils (Zhuang and Zhang, 2000), thus all heat Q was put on evaporation part as a linear heat flux during
AC
computation (Wen et al., 2005).
The Mohr-Coulomb yield criterion was adopted for the soils. The displacement in the x direction was constrained and that in the y direction was not. The bottom was taken as fixed and the top of the model was unconstrained. Because the soils above the spread footing move with the footing foundation, the interactions between the surrounding soils and the footing were ignored. In the model, the soils were considered to be expanded or shrunk with the change of soil temperature.
Due to construction disturbances and the high temperatures at the foundation surface, the permafrost under the footing may thaw in summer and re-freeze during winter (Li et al., 2014, this 14
ACCEPTED MANUSCRIPT special issue). The observational data showed that the disturbances during the construction resulted in the thaw of permafrost under the foundation and water tended to infiltrate into the
IP
T
foundation soils via refilled gravel (Li et al., 2014), which provides the necessary conditions for
SC R
the occurrence of frost heave and thaw settlement. The frost heave occurred when the soil was cooled, and the magnitude of the deformation depended on the frost heave coefficient and soil
NU
temperatures. The thaw settlement deformation was caused in a similar way.
Due to the lack of actual soil parameters along the QTPTL, especially the mechanical parameters,
MA
most of the mechanical parameters used in the simulation were chosen according to other
D
researchers’ experimental results (Wen et al., 2007). The experimental results of similar soil types
TE
and moisture contents were chosen and constituted the parameters used in the model. The parameters of active layer soil were based on silty clay experimental results, and the parameters of
CE P
permafrost soils were based on experiments on the icy clay with moderate liquid limits near the permafrost table (Wu and Ma, 1994; Liu, 2005). The parameters of thawed soils were obtained
AC
from routine soil experiments. Table 2 shows the mechanical parameters used in this computation. The elastic modulus of the concrete was taken as 31500 MPa, the concrete density was 2,500 kg·m−3, and Poisson's ratio was 0.17. To simulate the uplift and downward forces induced by wind, concentrated loads of -1000 kN and 1200 kN were exerted on the top of the foundation during freezing period (Ground surface temperature is colder than 0 C) and thawing period (Ground surface temperature is warmer than 0 C), respectively. MARC finite element program is employed to solve the problem. The thermal regime was calculated first by a finite element method. Then the results of the thermal regime were inputted into the deformation computation model as initial condition of each step of the deformation computation and the node temperatures 15
ACCEPTED MANUSCRIPT were inputted into the deformation computation model as a thermal load of each step of the deformation computation. Finally, the stress and strain on the foundation were calculated by a
IP
T
finite element method based on the deformation computation model. The frozen or thawed state of
SC R
the soils was determined by the node temperature, and the frost heave and thaw settlement were also judged by node temperature in the computation. The deformation amount due to frost heave and thaw settlement was ascertained based on changes in the node temperatures, and frost heave,
NU
thaw settlement, and compression coefficients. Thus, the thermal load in frozen soils was
MA
transformed into a displacement load, and the deformation process in the frost action can be simulated. The frost heave coefficient Br of the active layer and thawed soil under the foundation
D
was taken as 12% and the thaw settlement coefficient Ar of permafrost was taken as 30%
CE P
follows (Zhang, 2004):
TE
according to soil type and ice content. The compression coefficient Ay can be expressed as
Ay 1.2e3.3T
(22)
AC
Where T is permafrost temperature.
3.3.2 Computation results
Fig. 7 shows the thermal regime of simulated tower foundation. As shown in the Fig. 7(a), the thaw depth was approximately 4.0 m beneath the tower foundation on November 30, one year after construction. The thaw depth near the foundation was much bigger than that in natural ground. The ground temperature under the foundation was -0.5 C.
Fig. 7(b) shows the thermal regime of simulated tower foundation with thermosyphons on April 1,
16
ACCEPTED MANUSCRIPT one year after construction. The permafrost temperature under the foundation fell remarkably and a cold permafrost bulb with a temperature of -3 C formed due to the existence of thermosyphons.
IP
T
Compared with natural ground, the permafrost temperature at the same depth near the foundation
SC R
was lower. The results show that the thermosyphons had a clear cooling effect in cold seasons and soil refreezing was reinforced significantly.
NU
To evaluate the effect of Global warming, a warming rate of 2.6 ℃ in following 50 years was taken into consideration during computation. Fig. 7(c) shows the thermal regime of simulated
MA
tower foundation on November 30, 50th after construction. Permafrost table on natural ground
D
declines from 2.5 to about 3.3 m in the 50th year because of climate warming. Artificial
TE
permafrost table under the foundation declines to about 5.8 m and a huge thaw bulb has been formed beneath the foundation. In a word, ice-rich permafrost beneath the foundation has thawed
CE P
and big thaw settlement may occur, which endangers the stability of foundations significantly. Therefore, to keep the stability of the transmission line foundation, it is necessary to take an
AC
effective measure to counteract the influence of climate warming on permafrost under foundations.
Fig. 7(d) shows the thermal regime of simulated tower foundation with thermosyphons on April 1 and November 30, one year after construction. It can be seen from Fig. 7 (d) that permafrost under the foundation refreezes fully and the refreezing temperature of permafrost under the foundation is approximately -0.5℃. Due to global warming, permafrost under natural ground degrades seriously while there is a -0.5 ℃ permafrost bulb under the foundation. Calculated results show that the foundation with thermosyphons can compensate the influence of global warming and prevent
17
ACCEPTED MANUSCRIPT permafrost from thawing. The thermal stability of the foundation can be maintained in the service
T
time of 50 years if the remedial measure with thermosyphons only is carried out.
IP
Fig. 8 shows the ground surface deformation on November 30 and on April 1, one year after
SC R
construction. Numerical simulation results showed that deformation in the ground surface was very small in the thawing period. Due to soil compression, slight settlement occurred at the
NU
foundation and the deformation at the top of the foundation reached 0.06 m. In the freezing period, significant frost heave occurred. The amount of frost heave was the smallest at the foundation and
MA
it increased gradually with the increase in the distance from the foundation. The amount of frost
D
heave at 5 m from the foundation centerline reached 0.32 m. The heavy load of the foundation
TE
limited the occurrence of frost heave near the foundation and the amount of frost heaving at the top of the foundation is approximately 0.06 m. Fig. 9 shows the deformation at the top of the
CE P
foundation in the following 50 years. To simulate the foundation deformations in different MAGT permafrost regions, the MAGT of the upper boundary condition was set as -0.5 °C, -1.0 °C, and
AC
-1.5 °C. Numerical simulation results showed that continuous settlement occurred after the construction of the foundations and the amount of the deformation at the top of the foundation reached approximately 0.5 m in the 50th year. The deformations decreased with the decrease in ground temperature. The results with thermosyphons showed that slight frost heaving occurred in following 25 years and the deformation at the top of the foundation reached 0.08 m. In the 50th year, the deformation at the top of the foundation reached 0.1 m, which is significantly smaller than that without thermosyphon. Computational results indicated that the thermosyphons have a significant cooling effect on permafrost beneath the foundation and can significantly reduce the deformation at the top of foundation. 18
ACCEPTED MANUSCRIPT Fig. 10 shows the stress distribution of simulated tower foundation. Numerical simulation results on November 30, one year after construction showed that stress strengthened with depth and a
IP
T
stress bulb formed under the foundation due to the upper load. The stress under the foundation
SC R
was significantly greater than those at same deep under the ground surface, and it reached approximately 200 kPa. The stress under the stress bulb attenuated strongly with depth (Fig. 10(a)). Fig. 10(b) shows the stress distribution of simulated tower foundation on November 30, 50th year
NU
after construction. Comparison of Fig. 10(a) and 10(b) shows that the stress bulb in the 50th year
MA
was similar to that in the first year. Numerical results indicated that the stress distribution of
D
simulated tower foundation has not significant change in the following 50 years.
TE
Fig. 10(c) shows the stress distribution of simulated tower foundation on April 1, 50th year after construction. Comparison of Fig. 10(c) and 10(a) shows that the stress bulb in the freezing period
CE P
was significantly larger than that in the thawing period. The stress under the foundation reached approximately 900 kPa. The results indicated that the cooling effect of thermosyphons can result
AC
in additional stress under the foundation. The foundation can be jacked out of ground if the frost heave force increases or the resistance declines. The foundation suffered from enormous frost heave force, suggesting that the frost heave force induced by soil freezing can potentially threaten the safety and normal operation of the transmission line. Numerical simulation results showed that the additional stress beneath the tower foundation in the 50th year reduced significantly (Fig. 10(d)). Comparison of Figs. 10(c) and 10(d) indicates the stress beneath the foundation decreased from 900 to 300 kPa, suggesting that the normal frost heaving force beneath the foundation decreased significantly due to global warming.
19
ACCEPTED MANUSCRIPT 4 Conclusions
T
Based on the above analyses and discussions of the in-situ measurements and numerical
SC R
IP
simulation, it is concluded:
The contact stresses under transmission tower foundations without thermosyphons showed significant seasonal variations, depending nearby air or ground temperatures. The contact stresses
NU
showed significant daily variation, which was closely related to variations in air temperature.
MA
However, the wind speed did not have a significant impact on the contact stress. The observation data showed that the contact stresses with and without thermosyphons were free from the
TE
D
influences of wind.
If the global warming effect is considered in the model, a harmful deformation of the foundation
CE P
occurs and the settlement at the top of the foundation reached 0.5 m. Both the numerical simulation results and the observation data showed that the thermosyphons had a clear cooling
AC
effect in cold seasons and soil refreezing was reinforced significantly. The deformation of the foundation was reduced from 0.5 to 0.1 m due to the remedial measure with thermosyphons.
However, the cooling effect of thermosyphons led to the occurrence of frost heave and high contact stress. The stress beneath the foundation increased from 200 to 900 kPa. The occurrence of frost heave beneath the foundation increased the frost jacking risk of foundations. The global warming effect mitigated soil refreezing beneath the foundation and the stress beneath the foundation decreased from 900 kPa in the first year to 300 kPa in the 50th year.
These analyses may provide some useful insights into the possible mechanical states of the 20
ACCEPTED MANUSCRIPT QTPTL tower foundation-soil system for the design, construction, and operation of transmission
SC R
IP
T
lines in permafrost regions.
ACKNOWLEDGEMENTS
NU
The research project was supported by National Key Basic Research Program of China (Grant No. 2012CB026101), the National Natural Science Foundation of China (Grant No.41471061), the
MA
100 Talent Young Scientists Project of Chinese Academy of Sciences granted to Dr. Zhi Wen, the Science and Technology Project of the State Grid Corporation of China(Contact No. SGJSJS
TE
D
(2010)935-936), and the Fund of the State Key Laboratory of Frozen Soil Engineering (Grant Nos. SKLFSE-ZY-12 and SKLFSE-ZY-16).The authors sincerely thank the Nagqu Station of Plateau
CE P
Climate and Environment of Cold and Arid Regions Environmental and Engineering Research
AC
Institute for their meteorological data.
REFERENCES
Aksenov V.,Kistanov O., 2008. Estimation of resistance components to an axial load on piles embedded in permafrost. Soil Mechanics & Foundation Engineering, 45(2), 71-75. Cheng D.X., Zhang J.M., Liu H.J., Yu Q.H., Liu Z.W., 2009. The influence factor analysis for site seclect of transmission line in frozen earth area. Journal of Engineering Geology, 17(3): 329-333. (In Chinese)
21
ACCEPTED MANUSCRIPT Jiang H.P., Liu Z.R., 2006. ±500kVdirect current transmission line ground and foundation design in frozen ground area. Inner Mongolia Electric Power, 24(4), 14. (In Chinese)
IP
T
Johnson J. B., Esch D.C., 1985. Frost jacking forces on H and pipe pile embedded in Fairbanks silt.
SC R
In: Proceedings of the 4th International Symposium on Ground Freezing, Sapporo, Janpan, Volume 2, pp. 125-133.
Ladanyi B., Foriero A., 1998. Evolution of frost heaving stresses acting on a pile. In: Lewkowicz
MA
Yellowknife, N.W.T. Canada, pp. 623-633.
NU
A.G., Allard M.(eds.). Proceedings of the 7th International Conference on Permafrost,
Li G., Yu Q., Ma W., Mu Y., Li X., Chen Z., 2014. Laboratorytesting on heattransfer of frozen soil
D
blocks used as backfills of pile foundation in permafrost along Qinghai-Tibet electrical
TE
transmission line. Arabian Journal of Geosciences. DOI: 10.1007/s12517-014-1432-9.
CE P
Lyazgin A. L., Lyashenko V. S., Ostroborodov S. V., Ol’shanskii V. G., Bayasan R. M., Shevsov K. P., Pustovoit G. P., 2004. Experience in the prevention of frost heave of pipe foundations of
AC
transmission towers under northern conditions. Power Technology and Engineering, 38(2), 124-126.
Penner E., Irwin W. W., 1969. Adfreezing of leda clay to anchored footings columns. Canadian Geotechnical Journal, 6(3), 327-337. Penner E., Gold L. W., 1971. Transfer of heaving forces by adfreezing to columns and foundation walls in frost-susceptible soils. Canadian Geotechnical Journal, 8, 514-526. Penner E., 1974. Uplift forces on foundations in frost heaving soils. Canadian Geotechnical Journal, 11, 323-338. Perameswaran V. R., 1978. Adfreeze strength of frozen sand to model piles. Canada Geotechnique 22
ACCEPTED MANUSCRIPT Journal, 15(4), 494-500. Sadovsikiy A. V., 1973. Adfreeze between ground and foundation materials. In: Sanger F.j. (ed). National Academy of
IP
T
Proceedings of the 2nd International Conference on Permafrost,
SC R
Sciences,Washingtong D.C., pp. 650-653.
Saitykov, N.I., 1944. Calculating frost heaving forces on foundations. Izvestiia Akademia Nauk SSSR, Otdelenie Tekhnickeskikh Nauk, No. 6, 305-412
NU
Selvadurai A., Hu J., 1996. Axial loading of foundations embedded in frozen soil. International
MA
Journal of Offshore and Polar Engineering, 6(2), 650-653. Tong C., Guan F. 1985. Frost heaving and the prevention of freezing damage. Beijing, China
D
Water Power Press. (In Chinese)
TE
Trow, W.A., 1955. Frost action on small footings. Highway Research Board Bulletin No. 100,
CE P
22-27.
Wen Z., Sheng Y., Ma W., Qi J., Wu J., 2005. Analysis on effect of permafrost protection by
AC
two-phase closed thermosyphon and insulation jointly in permafrost regions. Cold Regions Science and Technology, 43(3), 150-163. Wen Z., Sheng Y., Jin H., Li S., Li G., Niu Y., 2010. Thermal Elasto-plastic computation Model of a buried oil pipeline subject to frost heave and thaw settlement. Cold Regions Science and Technology, 64(3), 248-255. Wang R., Wang W., Chen Y., 2005. Model experimental study on compressive bearing capacity of single pile in frozen soil. Journal of Glaciology and Geocryology, 27(2), 188193. (In Chinese) Wang X., Jiang D., Zhao X., 2005. Experimental study on bearing features of bored pile 23
ACCEPTED MANUSCRIPT undernon-refreezing condition in permafrost region, Chinese Journal of Geotechnical Engineering, 27(1), 81-84. (In Chinese)
IP
T
Wu Y.P., Sheng Y., Wang Y., Jin H.J., Chen W., 2010. Stresses and deformations in a buried oil
SC R
pipeline subject to differential frost heave in permafrost regions, Cold Regions Science and Technology, 64, 256-261.Xu G.F., Qi J. L., Jin H.J., 2010. Model test study on influence of freezing and thawing on the crude oil pipeline in cold regions. Cold Regions Science and
NU
Technology, 64, 262-270.
MA
Zhang J., Ma W., Wang D., Yuan X., 2008. In-situ experimental study of the bearing characteristics of cast-in-place pile in permafrost regions of the Tibetan Plateau. Chinese
D
Journal of Geotechnical Engineering, 30(3), 482-487. (In Chinese)
TE
Zhang J. M., 2004. Study on roadbed in permafrost regions on Qinghai-Tibetan Plateau and
CE P
classification of permafrost in highway engineering, Doctoral thesis of CAREERI, Chinese Academy of Sciences.
AC
Zhuang J., Zhang H., 2000. Heat pipe technology and engineering application. Chemistry Industry Press, Beijing. (In Chinese)
24
ACCEPTED MANUSCRIPT Table 1 Thermal properties of the materials in our computational domain C ( Jkg 1 C 1 )
(Wm1 C 1 ) (Wm1 C 1 )
Active layer soils
1,710
0.9
1.21
1.90
Ice-poor permafrost
1,680
1.1
1.39
Ice-rich permafrost
1,650
1.20
1.69
Concrete foundation
2,500
0.95
0.95
15
0.94
20
1.95
1.08
50
2.94
2.94
0
1.06
SC R
NU MA D TE CE P AC
25
w(%)
1.44
T
C ( Jkg 1 C 1 )
IP
(kgm3 )
Material type
ACCEPTED MANUSCRIPT Table 2 Mechanical properties of the materials in our computational domain
-10C
-5C
-2C
0C
ρ (kg·m-3)
1,920
1,920
1,920
1,920
1,920
1,920
Elastic modulus E (MPa)
200
100
50
23.4
6
6
Poisson’s ratio ν
0.32
0.32
0.32
0.32
0.35
0.35
Cohesion c(MPa)
0.6
0.6
0.6
0.57
0.15
0.15
Inter friction angle φ ()
26
26
26
26
24
24
ρ (kg·m-3)
1,834
1,834
1,834
1,834
1,834
1,834
300
100
70
3
3
0.15
0.15
0.15
0.15
0.2
0.2
1.3
1.3
1.3
1.3
0.1
0.2
Inter friction angle φ ()
20
20
20
20
18
18
ρ (kg·m-3)
2,500
2,500
2,500
2,500
2,500
2,500
Elastic modulus E (MPa)
31500
Poisson’s ratio ν
0.17
MA
layer
Elastic modulus E (MPa)
TE
Cohesion c(MPa)
Concrete
AC
CE P
Permafrost
500
D
Poisson’s ratio ν
IP
SC R
Active
26
T
-20C
NU
Material type
20C
31500
31500
31500
31500
31500
0.17
0.17
0.17
0.17
0.17
ACCEPTED MANUSCRIPT Uplift or downward loads
700
800
4100
Adfreeze bond
IP
200
T
Tangential frost heaving force
300
400
5300
Thermal resistor
Thin-film pressure sensor
Contact stress
SC R
1000 1000 3600
(b)
NU
(a)
Normal frost heaving force
Fig. 1 Sensor installation design at the test tower foundation (a) and force analysis (b)
AC
CE P
TE
D
MA
(Unit: mm)
27
ACCEPTED MANUSCRIPT 400 Sensor 1 Sensor 2 Sensor 3
350
T IP
250 200
SC R
Contact stress（kPa）
300
150
NU
100
0 1-May-11
1-Aug-11
1-Nov-11
MA
50
1-Feb-12
3-May-12
3-Aug-12
(a) 3-Nov-12
3-Feb-13
TE
D
Date
CE P
2000
1000
AC
Contact stress（kPa）
1500
Sensor 1 Sensor 2 Sensor 3
500
0 1-May-11
(b) 1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
28
3-Aug-12
3-Nov-12
3-Feb-13
ACCEPTED MANUSCRIPT
20
T SC R
IP
0
-10
-20
NU
Air temperature(℃)
10
1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
(c) 3-Aug-12
3-Nov-12
3-Feb-13
TE
D
-40 1-May-11
MA
-30
CE P
20
10
AC
Wind speed (m/s)
15
5
0 1-May-11
(d) 1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
3-Aug-12
3-Nov-12
3-Feb-13
Fig. 2 Variations of contact stress at the 490# tower foundation (a), Variations of contact stress at the 492# tower foundation (b), air temperature (c), and wind speed (d) 29
ACCEPTED MANUSCRIPT 0 490# tower foundation 492# tower foundation
T IP
-1
SC R
Ground temperature（℃）
-0.5
-1.5
1-Nov-11
3-May-12 Date
3-Aug-12
CE P
0.04
0.02
0
3-Nov-12
3-Feb-13
490# tower foundaiton 492# tower foundaiton
AC
Deformation at the top of foundation (m)
1-Feb-12
TE
0.06
-0.02
MA
1-Aug-11
(a)
D
-2.5 1-May-11
NU
-2
-0.04
-0.06
-0.08 1-May-11
(b) 1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
3-Aug-12
3-Nov-12
3-Feb-13
Fig. 3 Variations of ground temperature (a) and deformation (b) at the 490# and 492# tower foundations
30
ACCEPTED MANUSCRIPT 48
20
Wind speed Air temperature
16
-60
(a)
-100 10-Feb-12
MA
0 31-Jan-12
-80
D
05-Feb-12 Date
20
TE
500
10
CE P
400
0
Sensor 1 Sensor 3
AC
300
Sensor 2 Wind speed -10
100
-20
(b) 0 1-Jan-12
-30 11-Jan-12
21-Jan-12 Date
31
31-Jan-12
Wind speed（m/s）
8
Contact stress（kPa）
-40
SC R
24
200
-20
IP
32
Air temperaure（℃）
T
0
NU
Wind speed（m/s）
40
400
-20
Sensor 1 Sensor 3
-40
IP
300
Sensor 2 Air temperature
200
-80
0 1-Jan-12
11-Jan-12
MA
NU
100
-60
Air temperature（℃）
0
T
500
SC R
Contact stress（kPa）
ACCEPTED MANUSCRIPT
(c)
-100
21-Jan-12
31-Jan-12
Date
0
TE
D
500
200
-40 Sensor 1 Sensor 3
Sensor 2 Air temperature -60
Air temperature（℃）
CE P
300
100
-20
AC
Contact stress（kPa）
400
-80 (d)
0 21-Jan-12
-100 24-Jan-12
27-Jan-12
30-Jan-12
Date
Fig. 4 The relationship between daily variations in wind speed, air temperature and contact stress at 490# tower foundation, (a) daily variations in wind speed and in air temperature near the test tower foundations, (b) daily variations in wind speed and in the contact stress during January 2012, (c) daily variations in air temperature and in the contact stress during January 2012, (d) daily variations in air temperature and in the contact stress during the late January 2012 32
ACCEPTED MANUSCRIPT
20
IP
T
2700
(a)
11-Jan-12
2700
Sensor 3
-60
Wind speed
21-Jan-12
31-Jan-12
Date
0
-20
AC
2300
1900
-40
1500
-60 Sensor 1 Sensor 2 Sensor 3
1100
-80
Air temperature (b) 700 1-Jan-12
Wind speed（m/s）
Sensor 2
CE P
TE
-40 Sensor 1
-80
D
700 1-Jan-12
-20
-100 11-Jan-12
21-Jan-12 Date
33
31-Jan-12
Air temperature（℃）
1500
1100
Contact stress（kPa）
0
SC R NU
1900
MA
Contact stress（kPa）
2300
2300
-20
IP
1900
1500
NU
1100
700 21-Jan-12
23-Jan-12
MA
(c)
25-Jan-12
27-Jan-12
-40
-60 Sensor 1
Air temperature（℃）
0
T
2700
SC R
Contact stress（kPa）
ACCEPTED MANUSCRIPT
Sensor 2 Sensor 3
-80
Air temperature
-100 29-Jan-12
31-Jan-12
Date
D
Fig. 5 The relationship between daily variations in wind speed, air temperature and
TE
contact stress at 492# tower foundation, (a) daily variations in wind speed and in the
CE P
contact stress during January 2012, (b) daily variations in air temperature and in the contact stress during January 2012, (c) daily variations in air temperature and in the
AC
contact stress during the late January 2012
34
ACCEPTED MANUSCRIPT
T IP
Active layer
SC R
Ice-rich permafrost Concrete foundation
NU
Ice-poor permafrost
MA
Y (m)
X (m) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
AC
CE P
TE
D
Fig. 6 Geometry of axis-symmetrical footing foundation along QTPTL
35
ACCEPTED MANUSCRIPT 0
IP
T
-2
SC R
-6
NU
Depth (m)
-4
MA
-8
-10 0
3
6
(a) 9
12
15
TE
D
Width (m)
CE P
0
AC
-2
Depth (m)
-4
-6
-8
(b) -10 0
3
6
9 Width (m)
36
12
15
ACCEPTED MANUSCRIPT 0
IP
T
-2
SC R
-6
NU
Depth (m)
-4
MA
-8
-10 0
3
6
(c) 9
12
15
D
Width (m)
CE P
TE
0
-2
-6
AC
Depth (m)
-4
-8
-10
(d) 0
3
6
9
12
Width (m)
Fig. 7 Thermal regime of simulated tower foundation, (a) on November 30, one year after construction, (b) on April 1, one year after construction, (c) on November 30, 50th year after construction, (b) on April 1, 50th year after construction (Unit:C) 37
15
ACCEPTED MANUSCRIPT 0.35 0.30
T
0.15
IP
0.20
0.10
Ground surface deformation in thawing period
SC R
Ground surface deformation (m)
0.25
Ground surface deformation in freezing period 0.05
NU
0.00 -0.05
0
1
MA
-0.10
2 3 Distance from the central of the foundation (m)
4
5
0
10
15
Time（Year） 20
25
30
35
40
45
50
AC
Deformation at the top of foundation（m）
0.1
5
CE P
0
TE
D
Fig. 8 Ground surface deformation in freezing and thawing periods
-0.1
-0.2
-0.3
-0.4
-0.5
-0.5 ℃ -1.0 ℃ -1.5 ℃ -0.5 ℃ with thermosyphons -1.0 ℃ with thermosyphons -1.5 ℃ with thermosyphons
-0.6
Fig. 9 Predicted deformation at the top of simulated tower foundation in following 50 years 38
ACCEPTED MANUSCRIPT
T
0
SC R
IP
-1
NU
-2
MA D
-4
CE P
TE
-5
-6
-7
0
AC
Depth (m)
-3
1
(a) 2
3 Width (m)
39
4
5
ACCEPTED MANUSCRIPT 0
IP
T
-1
SC R
-2
NU MA
-4
TE
D
-5
CE P
-6
-7
0
AC
Depth (m)
-3
1
(b) 2
3 Width (m)
40
4
5
ACCEPTED MANUSCRIPT 0
SC R
IP
T
-1
-2
MA
Depth (m)
NU
-3
D
-4
-6
-7 0
AC
CE P
TE
-5
1
(c) 2
3 Width (m)
41
4
5
ACCEPTED MANUSCRIPT 0
IP
T
-1
SC R
-2
NU
-4
MA
Depth (m)
-3
TE
D
-5
-7
0
AC
CE P
-6
1
(b) 2
3
4
5
Width (m)
Fig. 10 Stress distribution of simulated tower foundation, (a) on November 30, one year after construction, (b) on November 30, 50th after construction, (c) on April 1, one year after construction, (d) on April 1, 50th year after construction (Unit: kPa)
42
ACCEPTED MANUSCRIPT Highlights
T
IP SC R NU MA D TE CE P
•
Ground temperature is the dominant factor that determines the stress variation. The refreezing of foundation soil results in significant increase in contact stress. Thaw settlement deformation may lead to harmful deformation of tower foundations.
AC
• •
43
S0165-232X(15)00123-8 doi: 10.1016/j.coldregions.2015.06.007 COLTEC 2118
To appear in:
Cold Regions Science and Technology
Received date: Revised date: Accepted date:
28 September 2014 2 June 2015 9 June 2015
Please cite this article as: Wen, Zhi, Yu, Qihao, Zhang, Mingli, Xue, Ke, Chen, Liangzhi, Li, Desheng, Stress and deformation characteristics of transmission tower foundations in permafrost regions along the Qinghai-Tibet Power Transmission Line, Cold Regions Science and Technology (2015), doi: 10.1016/j.coldregions.2015.06.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Stress and deformation characteristics of transmission tower foundations in permafrost
*
, QihaoYu, Mingli Zhang, Ke Xue, Liangzhi Chen, Desheng Li
SC R
IP
Zhi Wen
T
regions along the Qinghai-Tibet Power Transmission Line
(State Key Laboratory of Frozen Soil Engineering, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences,
NU
Lanzhou Gansu 730000, China)
MA
ABSTRACT: A transmission tower foundation embedded in frozen soil is subject to both the wind-induced uplift and frost heave forces. The frost heave results in an upward force acting on
TE
D
the foundation, while additional stress induced by the structure load may compress the underlying soils. The freezing- and thawing-induced deformations tend to cause further structural loads and
CE P
lead to instability problems within the structure. To evaluate the engineering risk and ensure the safety of the Qinghai-Tibet Power Transmission Line (QTPTL) system, stress sensors were
AC
installed at the bases of two test tower foundations to investigate the stress state of the tower foundations. Using data on air and ground temperatures, and the deformation of tower foundations, we analyzed the stress variation, and the causes were discussed here. The results showed that the stresses at the bases of tower foundations had a close relationship with air and ground temperatures. The cooling of the underlying soils led to the occurrence of frost heave, which pushed the foundations upward and caused a significant stress bulb under the bases of tower foundations. Seasonal variations in the contact stress depended on the seasonal freezing and
*
Corresponding author: Tel.: +0086-931-496-7299. E-mail address: [email protected](Z. Wen) 1
ACCEPTED MANUSCRIPT thawing of foundation soil. The contact stress increased with the cooling of the underlying soils and decreased with the warming of the underlying soils. The results also showed that the contact
IP
T
stress was free of the wind influence, i.e., the wind-induced uplift force was minor for the contact
SC R
stress. To fully understand the influences of freezing and thawing on the stress state of tower foundations, a thermal-elastico-plastic finite element model for the tower foundation-soil system was established, and the stresses and deformations of a tower foundation subject to frost heave and
NU
thaw settlement were simulated. The results showed that the frost heave force induced by soil
MA
freezing potentially threatens the safety and normal operation of the QTPTL. Thaw settlement may lead to harmful deformation of tower foundations if global warming is considered in the
D
numerical model. The remedial measure with thermosyphons only can reduce the settlement of the
TE
foundation and will increase the frost jacking risk of the foundation.
CE P
KEY WORDS: Qinghai-Tibetan Plateau, stress, tower foundation, frost heave force, transmission
AC
line, permafrost.
1 Introduction
The Qinghai-Tibet Power Transmission Line (QTPTL) runs across 1,038 km of permafrost and seasonally frozen ground in the Interior of the Qinghai-Tibetan Plateau. The mean annual air temperature of permafrost and seasonally frozen ground along the QTPTL varies between -3C and -7C and the minimum air temperature is lower than -37C in short durations. The active layer is subjected to annual freeze-thaw cycles and its thickness varies between 2 and 3 m. Thus, substantial heave force is expected due to the existence of extensive frost-susceptible soils and cold climate. Moreover, the wind-induced uplift force is another load for the transmission towers
2
ACCEPTED MANUSCRIPT and can reach more than 1,500 kN. Warm and ice-rich permafrost is widespread in permafrost regions of the Qinghai-Tibetan Plateau. The warming and subsequent thawing of ice-rich
IP
T
permafrost tends to result in serious engineering and environmental consequences. If not well
SC R
designed or built, the tower foundations may be jacked up in the freezing periods and subside in the thawing periods, which would result in expensive maintenance costs and significantly threatens the safety and operation of the transmission lines. In arctic and sub-arctic regions,
NU
intense frost-induced heaving has considerably deteriorated the condition of numerous power
MA
network objects and significantly threatens the construction and operation of transmission lines (e.g., Lyazgin et al. 2004; Jiang and Liu 2006; Cheng et al. 2009). Seasonal frost heave or thaw
D
settlement deformation can impact the mechanical state of the foundations, while the stress state
TE
of tower foundations and surrounding soils has a close relationship with foundation failures.
damage.
CE P
Therefore, it is important to understand how the stress changes in order to reduce tower foundation
AC
Previous studies of foundations in cold regions have been focused on the adfreeze bond strength and frost heave force (e.g., Penner, 1974; Tong and Guan, 1985; Johnson and Esch, 1985). Since the 1930s, measurements of the adfreeze strength between piles and frozen ground have been undertaken (Saitykov, 1944; Trow, 1955; Penner and Irwin, 1969; Penner and Gold, 1971; Sadovsikiy, 1973; Perameswaran, 1978; Ladanyi and Foriero, 1998). Many studies focused on the bearing capacity and uplift performance of piles in frozen soil in recent decades (Selvadurai and Hu, 1996; Wang R et al., 2005; Wang X et al., 2005; Aksenov and Kistanov, 2008; Zhang et al., 2008). To further understand the frozen soil-foundation interactions, researchers have used the finite element method and model testing to determine the distributive features of stress and 3
ACCEPTED MANUSCRIPT deformation. Wu et al. (2010) proposed a nonlinear elastico-plastic finite element model for a pipeline-soil system and calculated the mechanical performance of the oil pipeline subject to
IP
T
differential frost heave in permafrost regions. Xu et al. (2010) investigated the stress and strain of
SC R
a buried pipeline and surrounding soils by a model test, and analyzed the ambient temperature of the pipeline foundation, the displacement and axial strain, and the stress in the pipeline induced by frost heave and thaw settlement. However, the stress and deformation characteristics of tower
NU
foundations have received little attention and there has been limited effort towards the
MA
understanding of the mechanical behavior of foundations.
D
To investigate the stresses and deformation of tower foundations in the permafrost regions along
TE
the QTPTL, the thermal regime, stress state, and deformation were monitored and simulated. The results were analyzed to understand the frost damage processes of tower foundations, as well as to
AC
in cold regions.
CE P
make mitigative recommendations for the design and the construction of power transmission lines
2 On-site experimental investigations on stress and deformation of tower foundations
2.1 Site description and instrumentation
To study the stress and deformation of tower foundations in cold regions, two experimental tower foundations were constructed at Qingshui’he along the QTPTL in 2011. The elevation of the study site is 4,465 m a.s.l. and the average annual air temperature during the period of 2011 to 2012 at the site is -4.9 C, with extremes of approximately 19.97 and -38.04 C. Weather records show that the average annual wind speed is 4.8 m/s. Engineering geological
4
ACCEPTED MANUSCRIPT investigations at the site indicate that there is ice-rich clayey soil with a thickness of 0.5 to 4 m below the permafrost table and the ice content by volume is about 50%~80%. The mean
IP
T
annual ground temperature (MAGT) of the permafrost at the 490# test tower foundation is
SC R
-0.5 C and the permafrost table is 2.5 m. The MAGT of the permafrost at the 492# test tower foundation is -1.0 C and the permafrost table is 2.8 m. The thickness of ice-rich permafrost at 490# and 492# is 2.2 m and 3.0 m, respectively. The test tower foundations are located in a
NU
warm and ice-rich permafrost zone and have low thermal stability.
MA
The contact stresses at the soil-foundation interface are vital to understand the interaction between
D
structural foundations and the supporting soil. To investigate the contact stress characteristics of
TE
tower foundations, three thin-film pressure sensors were installed beneath each tower foundation. To monitor ground temperatures at the test foundation section, six boreholes were drilled in the
CE P
vicinity of the tower foundation soils (Fig. 1). The depths of these boreholes varied from 10 to 23 m. In each borehole, thermistors were installed at depth intervals of 0.5 m from the ground surface.
AC
The thermistors were made by the State Key Laboratory of Frozen Soil Engineering, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, and with a calibrated accuracy of ±0.05 C. The deformation of the tower foundations was measured by a water level instrument.
To study the cooling effect of two-phase closed thermosyphons, four thermosyphons were installed around the 492# tower foundation. The length of thermosyphons was 9 m; the evaporation part was 7 m-long and was embedded in the ground. The condensation part was 2 m-long and was exposed in the air. Fig. 1 shows the sensor installation design at the test footing
5
ACCEPTED MANUSCRIPT foundation and force analysis.
IP
T
3 Results and discussions
SC R
3.1 Seasonal variation of contact stress
Fig. 2a shows the seasonal variation of contact stress under the 490# test tower foundation. The
NU
contact stresses of 490# showed significant variations. The absolute value of contact stress decreased gradually after October and reached approximately 0 kPa in January. Then the contact
MA
stress increased gradually to approximately 300 kPa in May, and then it remained constant during the summer. The seasonal variation of contact stress had a close relationship with the seasonal
TE
D
variation in air temperature near the site (Fig. 2c). The contact stress decreased significantly with the decrease in air temperature in winter and increased with the increasing air temperature. The
CE P
contact stress showed a positive correlation with the seasonal variations of ground temperatures near the tower foundation (Fig. 3a). The tower foundation tended to suffer from the force induced
AC
by wind. However, our experimental results showed that the seasonal variations contact stress beneath the foundation did not appear to have a visible relationship with the wind speed near the site (Fig. 2d).
The 490# tower foundation is located in an ice-rich permafrost site with a MAGT of -0.5 C. The ground temperatures near the base of tower foundation show significant seasonal variation. The compression of warm permafrost leads to the occurrence of settlement deformation. Our monitoring showed that settlement deformation at 490# has occurred since 2010, and total settlement amount reached 0.08 m in the past three years (Fig. 3b). As shown in Fig. 1(b), the
6
ACCEPTED MANUSCRIPT contact stress depends on the load, frost heaving force, and adfreeze bond strength between the foundation surface and frozen ground. Sub-surface soil freezing causes the variation of tangential
IP
T
frost heaving force and adfreeze bond strength between the foundation surface and frozen ground.
SC R
The increase in tangential frost heaving force and adfreeze bond strength during soil freezing may offset the contact stress, which results in the decrease in contact stress in cold season. During thawing season, the disappearance of tangential frost heaving force and the decrease in adfreeze
NU
bond caused the increase in contact stress. Thus, the contact stress at 490# depends on the air
MA
temperature or ground temperature. Our monitoring showed that contact stress at 490# reached approximately 0 kPa in January, which implies that the tower foundation has a potential safety
D
hazard to be jacked out of the ground. The deformation observation showed that uplift does not
TE
occur due to the existence of adfreeze bond. To ensure the safety of the tower foundation, it is
measures.
CE P
necessary to continue the stress and deformation observations or to apply some anti-frost action
AC
Fig. 2(b) shows the seasonal variations in contact stress under the 492# test tower foundation. The value of contact stress at 492# is much higher than that at 490#. However, the seasonal variability of the contact stress at the tower 492# was much smaller than that at 490#. The contact stress increased linearly since May 2011 due to line hanging, and reached its peak in August 2011, after which it remained largely constant. In contrast to that at the tower 490#, the contact stress at 492# did not display significant seasonal variations and was free from the influence of air temperature and wind speed. Due to the cooling effect of thermosyphons, ground temperatures near the tower foundation clearly dropped since May 2011.The ground temperatures decreased from -0.5C in May 2011 to -2C in May 2013 (Fig. 3a). The monitoring of tower foundation deformation 7
ACCEPTED MANUSCRIPT showed that a slight frost heave occurred at 492#, and reached 0.02 m in the past three years (Fig.
T
3b).
IP
The 492# tower foundation is located in an ice-rich permafrost site with a MAGT of -1.0 C. The
SC R
abundant water at the site and the cooling effect of thermosyphons led to the occurrence of normal frost heaving force, which pushed the tower foundation upward. The resistance of load and
NU
adfreeze strength between the foundation surface and frozen ground limits the frost heave and
MA
causes high contact stress at 492# test foundation.
It is well known that contact stress depends on the load, wind-induced force, and frost heaving
D
force. The static load of the upper structure does not vary significantly with time. The
TE
wind-induced force includes uplift force and down force, depending on wind direction. In this
CE P
study area the wind direction and wind speed vary quickly during the day, but the annual variation of wind speed is not significant. Therefore, the wind did not have an obvious influence on the
AC
seasonal variations of the contact stress. The influence of frost heaving force on a tower foundation includes two parts: the tangential frost heaving force induced by soil freezing of the active layer, which partially offset the contact stress, and the normal frost heaving force induced by soil freezing under the base, which augments the contact stress.
3.2 Daily variations of contact stress
The contact stresses at 490# and 492# both showed significant daily variation. The relationships of the contact stress with air temperatures, ground temperatures, and wind speed were analyzed, respectively. Fig. 4(a) shows the daily variations in wind speed and air temperature near the site.
8
ACCEPTED MANUSCRIPT Observational data indicate that the variations in wind speed had a similar trend with that of air temperature in most instances. Wind speed increased gradually with the increase in air temperature
IP
T
during the day, whereas it decreased gradually with the drop of air temperature during the night. If
SC R
there is no disturbance of large-scale circulation, both wind speed and air temperature depend on the variation of solar radiation. Thus, in this study the daily variability in wind speed was similar
NU
to those in air temperatures in most cases.
Fig. 4(b) shows the daily variations in wind speed and in the contact stress at 490# during January
MA
2012. The results show that the wind speed in the first half of January 2012 was low but was
D
fluctuatingly greater in the second half. However, the daily variations of the contact stress were
TE
similar during the entire January, suggesting that contact stress was free from the influence of
stress.
CE P
wind speed. We therefore conclude that the wind speed had no significant impact on the contact
AC
Fig. 4(c) shows the daily variations in air temperature and in the contact stress at 490#. The results show that the daily variations in air temperatures had a close relationship with those of the contact stress. The contact stress responded quickly to the variations in air temperatures (Fig. 4(d)). The observational data at 492# also showed a similar rule (Fig. 5).It was deduced that the impact of air temperature on the contact stress may be related to the variations in tensile force caused by the thermal expansion and contraction of the transmission line system.
3.3 Numerical simulation analysis on thermal-mechanical process of tower foundations in permafrost regions
9
ACCEPTED MANUSCRIPT 3.3.1 Computation model and parameters
T
A thermal elasto-plastic computation model by Wen et al. (2010) was adopted to further study the
IP
thermal regime, stress variations and deformation characteristics of tower foundations in
SC R
permafrost regions.
3.3.1.1 Thermal regime computation model
NU
If the water infiltration into, and the water movement in the active layer are ignored, the heat
f
T T 1 T ( ) ( r ) t z z r r r
T f n
u
Tu Lw0 n t
T
c
the media;
(2）
u indicate
(3）
the frozen and thawed states of soils, respectively.
—specific heat capacity of the soil;
—temperature;
AC
conductivity;
and
f
CE P
—soil density;
TE
T f ( (t ), t ) Tu ( (t ), t ) Tm
where the subscripts
(1）
D
c
MA
conduction with phase change in frozen ground can be described as follows:
L —latent
r
and
z
—time,
—coefficient of heat
heat；w0 —the initial water content；Tm —frozen point of
— the frozen and thawed interface;
thawed interface; and,
t
n—
the direction vector of the frozen and
are the radial and vertical distance in cylindrical coordinates,
respectively. The conditions at the fixed boundaries are: T Ta
Where
Ta is
T a(Ta T ) n
soil temperature and
a is
(4）
a coefficient of heat convection between thermosyphon and
air. 3.3.1.2 Mechanics computation model 10
ACCEPTED MANUSCRIPT When the influence of soil temperature on soil deformation is considered, and the influence of soil
can be expressed as follows：
IP
Total strain
T
deformation on soil temperature is ignored, the governing equation can be expressed as follows
e p th ry f
SC R
(5）
Bu
strain;
B —the transformed matrix； —elasticity strain; e
T
ry —the
(7）
p —plastic
compression strain of thawing soil or frost heave strain;
MA
Where
NU
D
(6）
strain;
f
th —thermal
—the compression
strain of frozen soil; and D —total material factors matrix.
TE
D
1) Elastic strain
The elastic deformation obeys the general Hooke’s law:
D e
e 1
(8）
CE P
De e
(9）
AC
where： —stress matrix; e —strain matrix;
D —elasticity. e
For the plane strain problem, 1 E ( 1 ) [De ] (1 )(1 2 ) 1 0
1 1 0
0 1 2 2(1 ) 0
(10）
where E is elastic modulus, is Poisson’s ratio, and they both are functions of temperature T. 2) Plastic strain Plastic strain increment relates to not only to the stress increment, but also the stress state. The
plastic strain increment { ij p } can be expressed as follows： 11
ACCEPTED MANUSCRIPT
{ ij } p
Q ij
（11）
Adopting the associated flow law, [ D p ] can be expressed as follows：
T
S S 2 S3
S 32
（12）
Q Q y 1 x
S3
1 2 Q 2(1 ) xy
S0 S1
Q Q Q S2 S3 x y xy
（13） （14） （15） (16）
MA
S2
SC R
Q Q x 1 y
NU
S1
2 2
IP
S12 1 [D ] S1S 2 So S1S 3 p
Thus, the elasto-plastic matrix [ Dep ] can be expressed as follows：
D D D e
p
(17）
D
ep
CE P
3) Thermal strain
TE
where is flow parameter；Q is plastic function, and is shear stress.
A change in soil temperature will result in a soil volume change, and the expansion ratio or
AC
shrinkage ratio th can be expressed as follows:
th T
where
is linear expand coefficient ( 1 / C );
（18） T is current temperature (℃).
4) Thaw settlement and compression strain Frozen ground is a 4-component and 3-phase complex material composed of liquid water, ice, gas, and soil grains. It has special components, such as ice crystals, lens and ice interlayers. The ice will melt if the temperature exceeds the freezing point of soils, the thawing soil will consolidate and the soil grains will realign due to self-weight. The above mentioned properties of frozen soils are generally called thaw settlement and the deformation is generally called 12
ACCEPTED MANUSCRIPT thaw-settlement deformation. Thawing soils will be further compressed if a load is applied. The thaw settlement and compression of the thawed soils are generally expressed by thaw
IP
T
settlement and compression coefficients, which are related to the water content, dry density,
SC R
and soil type. The thaw settlement and compression strain ry can be expressed as follows: ry Ar Ay P
（19）
where Ar , Ay and P are thaw settlement coefficients (%), compression coefficient (MPa-1) and
NU
load (MPa), respectively.
MA
5) Frost heave strain
The impact of load on frost heave was ignored in the model due to small load. So the frost heave
D
is expressed by frost heave coefficients related to soil type and water supply condition. The frost
（20）
CE P
ry Br
TE
heave strain ry can be expressed as follows:
AC
where Br is frost heave coefficients (%).
3.3.2 Computation parameters
Fig. 6 shows the geometry of a typical axis-symmetrical footing foundation along the QTPTL. An active layer thickness of 2.5 m is assumed for a typical site. The site has ice-rich permafrost with a thickness of 5.4 m below the permafrost table and the moisture content is 50%. The ice-poor permafrost is located at a depth of 7.9 to 20.0 m. The thermal parameters, initial temperature conditions, and the finite element method used for simulating the thermal regimes were similar to those in Wen et al. (2005).The MAGT of the upper boundary condition was 1.0 °C, and the annual range of the ground surface temperature was 26 °C. The annual mean concrete surface 13
ACCEPTED MANUSCRIPT temperature of the upper boundary condition was set at 1.5 °C, and the annual range was 30 °C. The surface temperature varied in a modified attenuated sine wave. The lower boundary had a heat
IP
T
flux constant of 0.06 W·m−2. The vertical axis of the foundation had symmetrical boundary
SC R
condition, and the other side boundaries were adiabatic. Table 1 lists the thermal properties of the materials in the computational domain, in which the symbols “” and “+” denote frozen and the
NU
unfrozen states, respectively.
The thermal exchange of the thermosyphon is assumed to be:
MA
Q F T (t )
(21)
where Q is the heat exchange, T (t ) the temperature difference between environment and the
D
the coefficient of convection heat transfer, and F the effective area of the
TE
soil around,
radiator .The thermal resistance of the thermosyphon was not taken into account in computation
CE P
because it has a thermal conductivity capacity several orders higher than that of the soils (Zhuang and Zhang, 2000), thus all heat Q was put on evaporation part as a linear heat flux during
AC
computation (Wen et al., 2005).
The Mohr-Coulomb yield criterion was adopted for the soils. The displacement in the x direction was constrained and that in the y direction was not. The bottom was taken as fixed and the top of the model was unconstrained. Because the soils above the spread footing move with the footing foundation, the interactions between the surrounding soils and the footing were ignored. In the model, the soils were considered to be expanded or shrunk with the change of soil temperature.
Due to construction disturbances and the high temperatures at the foundation surface, the permafrost under the footing may thaw in summer and re-freeze during winter (Li et al., 2014, this 14
ACCEPTED MANUSCRIPT special issue). The observational data showed that the disturbances during the construction resulted in the thaw of permafrost under the foundation and water tended to infiltrate into the
IP
T
foundation soils via refilled gravel (Li et al., 2014), which provides the necessary conditions for
SC R
the occurrence of frost heave and thaw settlement. The frost heave occurred when the soil was cooled, and the magnitude of the deformation depended on the frost heave coefficient and soil
NU
temperatures. The thaw settlement deformation was caused in a similar way.
Due to the lack of actual soil parameters along the QTPTL, especially the mechanical parameters,
MA
most of the mechanical parameters used in the simulation were chosen according to other
D
researchers’ experimental results (Wen et al., 2007). The experimental results of similar soil types
TE
and moisture contents were chosen and constituted the parameters used in the model. The parameters of active layer soil were based on silty clay experimental results, and the parameters of
CE P
permafrost soils were based on experiments on the icy clay with moderate liquid limits near the permafrost table (Wu and Ma, 1994; Liu, 2005). The parameters of thawed soils were obtained
AC
from routine soil experiments. Table 2 shows the mechanical parameters used in this computation. The elastic modulus of the concrete was taken as 31500 MPa, the concrete density was 2,500 kg·m−3, and Poisson's ratio was 0.17. To simulate the uplift and downward forces induced by wind, concentrated loads of -1000 kN and 1200 kN were exerted on the top of the foundation during freezing period (Ground surface temperature is colder than 0 C) and thawing period (Ground surface temperature is warmer than 0 C), respectively. MARC finite element program is employed to solve the problem. The thermal regime was calculated first by a finite element method. Then the results of the thermal regime were inputted into the deformation computation model as initial condition of each step of the deformation computation and the node temperatures 15
ACCEPTED MANUSCRIPT were inputted into the deformation computation model as a thermal load of each step of the deformation computation. Finally, the stress and strain on the foundation were calculated by a
IP
T
finite element method based on the deformation computation model. The frozen or thawed state of
SC R
the soils was determined by the node temperature, and the frost heave and thaw settlement were also judged by node temperature in the computation. The deformation amount due to frost heave and thaw settlement was ascertained based on changes in the node temperatures, and frost heave,
NU
thaw settlement, and compression coefficients. Thus, the thermal load in frozen soils was
MA
transformed into a displacement load, and the deformation process in the frost action can be simulated. The frost heave coefficient Br of the active layer and thawed soil under the foundation
D
was taken as 12% and the thaw settlement coefficient Ar of permafrost was taken as 30%
CE P
follows (Zhang, 2004):
TE
according to soil type and ice content. The compression coefficient Ay can be expressed as
Ay 1.2e3.3T
(22)
AC
Where T is permafrost temperature.
3.3.2 Computation results
Fig. 7 shows the thermal regime of simulated tower foundation. As shown in the Fig. 7(a), the thaw depth was approximately 4.0 m beneath the tower foundation on November 30, one year after construction. The thaw depth near the foundation was much bigger than that in natural ground. The ground temperature under the foundation was -0.5 C.
Fig. 7(b) shows the thermal regime of simulated tower foundation with thermosyphons on April 1,
16
ACCEPTED MANUSCRIPT one year after construction. The permafrost temperature under the foundation fell remarkably and a cold permafrost bulb with a temperature of -3 C formed due to the existence of thermosyphons.
IP
T
Compared with natural ground, the permafrost temperature at the same depth near the foundation
SC R
was lower. The results show that the thermosyphons had a clear cooling effect in cold seasons and soil refreezing was reinforced significantly.
NU
To evaluate the effect of Global warming, a warming rate of 2.6 ℃ in following 50 years was taken into consideration during computation. Fig. 7(c) shows the thermal regime of simulated
MA
tower foundation on November 30, 50th after construction. Permafrost table on natural ground
D
declines from 2.5 to about 3.3 m in the 50th year because of climate warming. Artificial
TE
permafrost table under the foundation declines to about 5.8 m and a huge thaw bulb has been formed beneath the foundation. In a word, ice-rich permafrost beneath the foundation has thawed
CE P
and big thaw settlement may occur, which endangers the stability of foundations significantly. Therefore, to keep the stability of the transmission line foundation, it is necessary to take an
AC
effective measure to counteract the influence of climate warming on permafrost under foundations.
Fig. 7(d) shows the thermal regime of simulated tower foundation with thermosyphons on April 1 and November 30, one year after construction. It can be seen from Fig. 7 (d) that permafrost under the foundation refreezes fully and the refreezing temperature of permafrost under the foundation is approximately -0.5℃. Due to global warming, permafrost under natural ground degrades seriously while there is a -0.5 ℃ permafrost bulb under the foundation. Calculated results show that the foundation with thermosyphons can compensate the influence of global warming and prevent
17
ACCEPTED MANUSCRIPT permafrost from thawing. The thermal stability of the foundation can be maintained in the service
T
time of 50 years if the remedial measure with thermosyphons only is carried out.
IP
Fig. 8 shows the ground surface deformation on November 30 and on April 1, one year after
SC R
construction. Numerical simulation results showed that deformation in the ground surface was very small in the thawing period. Due to soil compression, slight settlement occurred at the
NU
foundation and the deformation at the top of the foundation reached 0.06 m. In the freezing period, significant frost heave occurred. The amount of frost heave was the smallest at the foundation and
MA
it increased gradually with the increase in the distance from the foundation. The amount of frost
D
heave at 5 m from the foundation centerline reached 0.32 m. The heavy load of the foundation
TE
limited the occurrence of frost heave near the foundation and the amount of frost heaving at the top of the foundation is approximately 0.06 m. Fig. 9 shows the deformation at the top of the
CE P
foundation in the following 50 years. To simulate the foundation deformations in different MAGT permafrost regions, the MAGT of the upper boundary condition was set as -0.5 °C, -1.0 °C, and
AC
-1.5 °C. Numerical simulation results showed that continuous settlement occurred after the construction of the foundations and the amount of the deformation at the top of the foundation reached approximately 0.5 m in the 50th year. The deformations decreased with the decrease in ground temperature. The results with thermosyphons showed that slight frost heaving occurred in following 25 years and the deformation at the top of the foundation reached 0.08 m. In the 50th year, the deformation at the top of the foundation reached 0.1 m, which is significantly smaller than that without thermosyphon. Computational results indicated that the thermosyphons have a significant cooling effect on permafrost beneath the foundation and can significantly reduce the deformation at the top of foundation. 18
ACCEPTED MANUSCRIPT Fig. 10 shows the stress distribution of simulated tower foundation. Numerical simulation results on November 30, one year after construction showed that stress strengthened with depth and a
IP
T
stress bulb formed under the foundation due to the upper load. The stress under the foundation
SC R
was significantly greater than those at same deep under the ground surface, and it reached approximately 200 kPa. The stress under the stress bulb attenuated strongly with depth (Fig. 10(a)). Fig. 10(b) shows the stress distribution of simulated tower foundation on November 30, 50th year
NU
after construction. Comparison of Fig. 10(a) and 10(b) shows that the stress bulb in the 50th year
MA
was similar to that in the first year. Numerical results indicated that the stress distribution of
D
simulated tower foundation has not significant change in the following 50 years.
TE
Fig. 10(c) shows the stress distribution of simulated tower foundation on April 1, 50th year after construction. Comparison of Fig. 10(c) and 10(a) shows that the stress bulb in the freezing period
CE P
was significantly larger than that in the thawing period. The stress under the foundation reached approximately 900 kPa. The results indicated that the cooling effect of thermosyphons can result
AC
in additional stress under the foundation. The foundation can be jacked out of ground if the frost heave force increases or the resistance declines. The foundation suffered from enormous frost heave force, suggesting that the frost heave force induced by soil freezing can potentially threaten the safety and normal operation of the transmission line. Numerical simulation results showed that the additional stress beneath the tower foundation in the 50th year reduced significantly (Fig. 10(d)). Comparison of Figs. 10(c) and 10(d) indicates the stress beneath the foundation decreased from 900 to 300 kPa, suggesting that the normal frost heaving force beneath the foundation decreased significantly due to global warming.
19
ACCEPTED MANUSCRIPT 4 Conclusions
T
Based on the above analyses and discussions of the in-situ measurements and numerical
SC R
IP
simulation, it is concluded:
The contact stresses under transmission tower foundations without thermosyphons showed significant seasonal variations, depending nearby air or ground temperatures. The contact stresses
NU
showed significant daily variation, which was closely related to variations in air temperature.
MA
However, the wind speed did not have a significant impact on the contact stress. The observation data showed that the contact stresses with and without thermosyphons were free from the
TE
D
influences of wind.
If the global warming effect is considered in the model, a harmful deformation of the foundation
CE P
occurs and the settlement at the top of the foundation reached 0.5 m. Both the numerical simulation results and the observation data showed that the thermosyphons had a clear cooling
AC
effect in cold seasons and soil refreezing was reinforced significantly. The deformation of the foundation was reduced from 0.5 to 0.1 m due to the remedial measure with thermosyphons.
However, the cooling effect of thermosyphons led to the occurrence of frost heave and high contact stress. The stress beneath the foundation increased from 200 to 900 kPa. The occurrence of frost heave beneath the foundation increased the frost jacking risk of foundations. The global warming effect mitigated soil refreezing beneath the foundation and the stress beneath the foundation decreased from 900 kPa in the first year to 300 kPa in the 50th year.
These analyses may provide some useful insights into the possible mechanical states of the 20
ACCEPTED MANUSCRIPT QTPTL tower foundation-soil system for the design, construction, and operation of transmission
SC R
IP
T
lines in permafrost regions.
ACKNOWLEDGEMENTS
NU
The research project was supported by National Key Basic Research Program of China (Grant No. 2012CB026101), the National Natural Science Foundation of China (Grant No.41471061), the
MA
100 Talent Young Scientists Project of Chinese Academy of Sciences granted to Dr. Zhi Wen, the Science and Technology Project of the State Grid Corporation of China(Contact No. SGJSJS
TE
D
(2010)935-936), and the Fund of the State Key Laboratory of Frozen Soil Engineering (Grant Nos. SKLFSE-ZY-12 and SKLFSE-ZY-16).The authors sincerely thank the Nagqu Station of Plateau
CE P
Climate and Environment of Cold and Arid Regions Environmental and Engineering Research
AC
Institute for their meteorological data.
REFERENCES
Aksenov V.,Kistanov O., 2008. Estimation of resistance components to an axial load on piles embedded in permafrost. Soil Mechanics & Foundation Engineering, 45(2), 71-75. Cheng D.X., Zhang J.M., Liu H.J., Yu Q.H., Liu Z.W., 2009. The influence factor analysis for site seclect of transmission line in frozen earth area. Journal of Engineering Geology, 17(3): 329-333. (In Chinese)
21
ACCEPTED MANUSCRIPT Jiang H.P., Liu Z.R., 2006. ±500kVdirect current transmission line ground and foundation design in frozen ground area. Inner Mongolia Electric Power, 24(4), 14. (In Chinese)
IP
T
Johnson J. B., Esch D.C., 1985. Frost jacking forces on H and pipe pile embedded in Fairbanks silt.
SC R
In: Proceedings of the 4th International Symposium on Ground Freezing, Sapporo, Janpan, Volume 2, pp. 125-133.
Ladanyi B., Foriero A., 1998. Evolution of frost heaving stresses acting on a pile. In: Lewkowicz
MA
Yellowknife, N.W.T. Canada, pp. 623-633.
NU
A.G., Allard M.(eds.). Proceedings of the 7th International Conference on Permafrost,
Li G., Yu Q., Ma W., Mu Y., Li X., Chen Z., 2014. Laboratorytesting on heattransfer of frozen soil
D
blocks used as backfills of pile foundation in permafrost along Qinghai-Tibet electrical
TE
transmission line. Arabian Journal of Geosciences. DOI: 10.1007/s12517-014-1432-9.
CE P
Lyazgin A. L., Lyashenko V. S., Ostroborodov S. V., Ol’shanskii V. G., Bayasan R. M., Shevsov K. P., Pustovoit G. P., 2004. Experience in the prevention of frost heave of pipe foundations of
AC
transmission towers under northern conditions. Power Technology and Engineering, 38(2), 124-126.
Penner E., Irwin W. W., 1969. Adfreezing of leda clay to anchored footings columns. Canadian Geotechnical Journal, 6(3), 327-337. Penner E., Gold L. W., 1971. Transfer of heaving forces by adfreezing to columns and foundation walls in frost-susceptible soils. Canadian Geotechnical Journal, 8, 514-526. Penner E., 1974. Uplift forces on foundations in frost heaving soils. Canadian Geotechnical Journal, 11, 323-338. Perameswaran V. R., 1978. Adfreeze strength of frozen sand to model piles. Canada Geotechnique 22
ACCEPTED MANUSCRIPT Journal, 15(4), 494-500. Sadovsikiy A. V., 1973. Adfreeze between ground and foundation materials. In: Sanger F.j. (ed). National Academy of
IP
T
Proceedings of the 2nd International Conference on Permafrost,
SC R
Sciences,Washingtong D.C., pp. 650-653.
Saitykov, N.I., 1944. Calculating frost heaving forces on foundations. Izvestiia Akademia Nauk SSSR, Otdelenie Tekhnickeskikh Nauk, No. 6, 305-412
NU
Selvadurai A., Hu J., 1996. Axial loading of foundations embedded in frozen soil. International
MA
Journal of Offshore and Polar Engineering, 6(2), 650-653. Tong C., Guan F. 1985. Frost heaving and the prevention of freezing damage. Beijing, China
D
Water Power Press. (In Chinese)
TE
Trow, W.A., 1955. Frost action on small footings. Highway Research Board Bulletin No. 100,
CE P
22-27.
Wen Z., Sheng Y., Ma W., Qi J., Wu J., 2005. Analysis on effect of permafrost protection by
AC
two-phase closed thermosyphon and insulation jointly in permafrost regions. Cold Regions Science and Technology, 43(3), 150-163. Wen Z., Sheng Y., Jin H., Li S., Li G., Niu Y., 2010. Thermal Elasto-plastic computation Model of a buried oil pipeline subject to frost heave and thaw settlement. Cold Regions Science and Technology, 64(3), 248-255. Wang R., Wang W., Chen Y., 2005. Model experimental study on compressive bearing capacity of single pile in frozen soil. Journal of Glaciology and Geocryology, 27(2), 188193. (In Chinese) Wang X., Jiang D., Zhao X., 2005. Experimental study on bearing features of bored pile 23
ACCEPTED MANUSCRIPT undernon-refreezing condition in permafrost region, Chinese Journal of Geotechnical Engineering, 27(1), 81-84. (In Chinese)
IP
T
Wu Y.P., Sheng Y., Wang Y., Jin H.J., Chen W., 2010. Stresses and deformations in a buried oil
SC R
pipeline subject to differential frost heave in permafrost regions, Cold Regions Science and Technology, 64, 256-261.Xu G.F., Qi J. L., Jin H.J., 2010. Model test study on influence of freezing and thawing on the crude oil pipeline in cold regions. Cold Regions Science and
NU
Technology, 64, 262-270.
MA
Zhang J., Ma W., Wang D., Yuan X., 2008. In-situ experimental study of the bearing characteristics of cast-in-place pile in permafrost regions of the Tibetan Plateau. Chinese
D
Journal of Geotechnical Engineering, 30(3), 482-487. (In Chinese)
TE
Zhang J. M., 2004. Study on roadbed in permafrost regions on Qinghai-Tibetan Plateau and
CE P
classification of permafrost in highway engineering, Doctoral thesis of CAREERI, Chinese Academy of Sciences.
AC
Zhuang J., Zhang H., 2000. Heat pipe technology and engineering application. Chemistry Industry Press, Beijing. (In Chinese)
24
ACCEPTED MANUSCRIPT Table 1 Thermal properties of the materials in our computational domain C ( Jkg 1 C 1 )
(Wm1 C 1 ) (Wm1 C 1 )
Active layer soils
1,710
0.9
1.21
1.90
Ice-poor permafrost
1,680
1.1
1.39
Ice-rich permafrost
1,650
1.20
1.69
Concrete foundation
2,500
0.95
0.95
15
0.94
20
1.95
1.08
50
2.94
2.94
0
1.06
SC R
NU MA D TE CE P AC
25
w(%)
1.44
T
C ( Jkg 1 C 1 )
IP
(kgm3 )
Material type
ACCEPTED MANUSCRIPT Table 2 Mechanical properties of the materials in our computational domain
-10C
-5C
-2C
0C
ρ (kg·m-3)
1,920
1,920
1,920
1,920
1,920
1,920
Elastic modulus E (MPa)
200
100
50
23.4
6
6
Poisson’s ratio ν
0.32
0.32
0.32
0.32
0.35
0.35
Cohesion c(MPa)
0.6
0.6
0.6
0.57
0.15
0.15
Inter friction angle φ ()
26
26
26
26
24
24
ρ (kg·m-3)
1,834
1,834
1,834
1,834
1,834
1,834
300
100
70
3
3
0.15
0.15
0.15
0.15
0.2
0.2
1.3
1.3
1.3
1.3
0.1
0.2
Inter friction angle φ ()
20
20
20
20
18
18
ρ (kg·m-3)
2,500
2,500
2,500
2,500
2,500
2,500
Elastic modulus E (MPa)
31500
Poisson’s ratio ν
0.17
MA
layer
Elastic modulus E (MPa)
TE
Cohesion c(MPa)
Concrete
AC
CE P
Permafrost
500
D
Poisson’s ratio ν
IP
SC R
Active
26
T
-20C
NU
Material type
20C
31500
31500
31500
31500
31500
0.17
0.17
0.17
0.17
0.17
ACCEPTED MANUSCRIPT Uplift or downward loads
700
800
4100
Adfreeze bond
IP
200
T
Tangential frost heaving force
300
400
5300
Thermal resistor
Thin-film pressure sensor
Contact stress
SC R
1000 1000 3600
(b)
NU
(a)
Normal frost heaving force
Fig. 1 Sensor installation design at the test tower foundation (a) and force analysis (b)
AC
CE P
TE
D
MA
(Unit: mm)
27
ACCEPTED MANUSCRIPT 400 Sensor 1 Sensor 2 Sensor 3
350
T IP
250 200
SC R
Contact stress（kPa）
300
150
NU
100
0 1-May-11
1-Aug-11
1-Nov-11
MA
50
1-Feb-12
3-May-12
3-Aug-12
(a) 3-Nov-12
3-Feb-13
TE
D
Date
CE P
2000
1000
AC
Contact stress（kPa）
1500
Sensor 1 Sensor 2 Sensor 3
500
0 1-May-11
(b) 1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
28
3-Aug-12
3-Nov-12
3-Feb-13
ACCEPTED MANUSCRIPT
20
T SC R
IP
0
-10
-20
NU
Air temperature(℃)
10
1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
(c) 3-Aug-12
3-Nov-12
3-Feb-13
TE
D
-40 1-May-11
MA
-30
CE P
20
10
AC
Wind speed (m/s)
15
5
0 1-May-11
(d) 1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
3-Aug-12
3-Nov-12
3-Feb-13
Fig. 2 Variations of contact stress at the 490# tower foundation (a), Variations of contact stress at the 492# tower foundation (b), air temperature (c), and wind speed (d) 29
ACCEPTED MANUSCRIPT 0 490# tower foundation 492# tower foundation
T IP
-1
SC R
Ground temperature（℃）
-0.5
-1.5
1-Nov-11
3-May-12 Date
3-Aug-12
CE P
0.04
0.02
0
3-Nov-12
3-Feb-13
490# tower foundaiton 492# tower foundaiton
AC
Deformation at the top of foundation (m)
1-Feb-12
TE
0.06
-0.02
MA
1-Aug-11
(a)
D
-2.5 1-May-11
NU
-2
-0.04
-0.06
-0.08 1-May-11
(b) 1-Aug-11
1-Nov-11
1-Feb-12
3-May-12 Date
3-Aug-12
3-Nov-12
3-Feb-13
Fig. 3 Variations of ground temperature (a) and deformation (b) at the 490# and 492# tower foundations
30
ACCEPTED MANUSCRIPT 48
20
Wind speed Air temperature
16
-60
(a)
-100 10-Feb-12
MA
0 31-Jan-12
-80
D
05-Feb-12 Date
20
TE
500
10
CE P
400
0
Sensor 1 Sensor 3
AC
300
Sensor 2 Wind speed -10
100
-20
(b) 0 1-Jan-12
-30 11-Jan-12
21-Jan-12 Date
31
31-Jan-12
Wind speed（m/s）
8
Contact stress（kPa）
-40
SC R
24
200
-20
IP
32
Air temperaure（℃）
T
0
NU
Wind speed（m/s）
40
400
-20
Sensor 1 Sensor 3
-40
IP
300
Sensor 2 Air temperature
200
-80
0 1-Jan-12
11-Jan-12
MA
NU
100
-60
Air temperature（℃）
0
T
500
SC R
Contact stress（kPa）
ACCEPTED MANUSCRIPT
(c)
-100
21-Jan-12
31-Jan-12
Date
0
TE
D
500
200
-40 Sensor 1 Sensor 3
Sensor 2 Air temperature -60
Air temperature（℃）
CE P
300
100
-20
AC
Contact stress（kPa）
400
-80 (d)
0 21-Jan-12
-100 24-Jan-12
27-Jan-12
30-Jan-12
Date
Fig. 4 The relationship between daily variations in wind speed, air temperature and contact stress at 490# tower foundation, (a) daily variations in wind speed and in air temperature near the test tower foundations, (b) daily variations in wind speed and in the contact stress during January 2012, (c) daily variations in air temperature and in the contact stress during January 2012, (d) daily variations in air temperature and in the contact stress during the late January 2012 32
ACCEPTED MANUSCRIPT
20
IP
T
2700
(a)
11-Jan-12
2700
Sensor 3
-60
Wind speed
21-Jan-12
31-Jan-12
Date
0
-20
AC
2300
1900
-40
1500
-60 Sensor 1 Sensor 2 Sensor 3
1100
-80
Air temperature (b) 700 1-Jan-12
Wind speed（m/s）
Sensor 2
CE P
TE
-40 Sensor 1
-80
D
700 1-Jan-12
-20
-100 11-Jan-12
21-Jan-12 Date
33
31-Jan-12
Air temperature（℃）
1500
1100
Contact stress（kPa）
0
SC R NU
1900
MA
Contact stress（kPa）
2300
2300
-20
IP
1900
1500
NU
1100
700 21-Jan-12
23-Jan-12
MA
(c)
25-Jan-12
27-Jan-12
-40
-60 Sensor 1
Air temperature（℃）
0
T
2700
SC R
Contact stress（kPa）
ACCEPTED MANUSCRIPT
Sensor 2 Sensor 3
-80
Air temperature
-100 29-Jan-12
31-Jan-12
Date
D
Fig. 5 The relationship between daily variations in wind speed, air temperature and
TE
contact stress at 492# tower foundation, (a) daily variations in wind speed and in the
CE P
contact stress during January 2012, (b) daily variations in air temperature and in the contact stress during January 2012, (c) daily variations in air temperature and in the
AC
contact stress during the late January 2012
34
ACCEPTED MANUSCRIPT
T IP
Active layer
SC R
Ice-rich permafrost Concrete foundation
NU
Ice-poor permafrost
MA
Y (m)
X (m) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
AC
CE P
TE
D
Fig. 6 Geometry of axis-symmetrical footing foundation along QTPTL
35
ACCEPTED MANUSCRIPT 0
IP
T
-2
SC R
-6
NU
Depth (m)
-4
MA
-8
-10 0
3
6
(a) 9
12
15
TE
D
Width (m)
CE P
0
AC
-2
Depth (m)
-4
-6
-8
(b) -10 0
3
6
9 Width (m)
36
12
15
ACCEPTED MANUSCRIPT 0
IP
T
-2
SC R
-6
NU
Depth (m)
-4
MA
-8
-10 0
3
6
(c) 9
12
15
D
Width (m)
CE P
TE
0
-2
-6
AC
Depth (m)
-4
-8
-10
(d) 0
3
6
9
12
Width (m)
Fig. 7 Thermal regime of simulated tower foundation, (a) on November 30, one year after construction, (b) on April 1, one year after construction, (c) on November 30, 50th year after construction, (b) on April 1, 50th year after construction (Unit:C) 37
15
ACCEPTED MANUSCRIPT 0.35 0.30
T
0.15
IP
0.20
0.10
Ground surface deformation in thawing period
SC R
Ground surface deformation (m)
0.25
Ground surface deformation in freezing period 0.05
NU
0.00 -0.05
0
1
MA
-0.10
2 3 Distance from the central of the foundation (m)
4
5
0
10
15
Time（Year） 20
25
30
35
40
45
50
AC
Deformation at the top of foundation（m）
0.1
5
CE P
0
TE
D
Fig. 8 Ground surface deformation in freezing and thawing periods
-0.1
-0.2
-0.3
-0.4
-0.5
-0.5 ℃ -1.0 ℃ -1.5 ℃ -0.5 ℃ with thermosyphons -1.0 ℃ with thermosyphons -1.5 ℃ with thermosyphons
-0.6
Fig. 9 Predicted deformation at the top of simulated tower foundation in following 50 years 38
ACCEPTED MANUSCRIPT
T
0
SC R
IP
-1
NU
-2
MA D
-4
CE P
TE
-5
-6
-7
0
AC
Depth (m)
-3
1
(a) 2
3 Width (m)
39
4
5
ACCEPTED MANUSCRIPT 0
IP
T
-1
SC R
-2
NU MA
-4
TE
D
-5
CE P
-6
-7
0
AC
Depth (m)
-3
1
(b) 2
3 Width (m)
40
4
5
ACCEPTED MANUSCRIPT 0
SC R
IP
T
-1
-2
MA
Depth (m)
NU
-3
D
-4
-6
-7 0
AC
CE P
TE
-5
1
(c) 2
3 Width (m)
41
4
5
ACCEPTED MANUSCRIPT 0
IP
T
-1
SC R
-2
NU
-4
MA
Depth (m)
-3
TE
D
-5
-7
0
AC
CE P
-6
1
(b) 2
3
4
5
Width (m)
Fig. 10 Stress distribution of simulated tower foundation, (a) on November 30, one year after construction, (b) on November 30, 50th after construction, (c) on April 1, one year after construction, (d) on April 1, 50th year after construction (Unit: kPa)
42
ACCEPTED MANUSCRIPT Highlights
T
IP SC R NU MA D TE CE P
•
Ground temperature is the dominant factor that determines the stress variation. The refreezing of foundation soil results in significant increase in contact stress. Thaw settlement deformation may lead to harmful deformation of tower foundations.
AC
• •
43