JianKun Liu , TengFei Wang , Zhi Wen

1. Research Institute of Subgrade and Foundation Engineering, School of Civil Engineering, Beijing Jiaotong University,Beijing 100044, China

2. Department of Civil and Environmental Engineering, University of Wisconsin-Madison, Madison WI 53706, United States

3. Key Laboratory of Highway Construction and Maintenance Technology in Permafrost Region, Qinghai Research Institute of Transportation, Xining, Qinghai 810000, China

4. State Key Laboratory of Frozen Soil Engineering, Northwest Institute of Eco-Environment and Resources, Chinese Academy Sciences, Lanzhou, Gansu 730000, China

1　Introduction

Pile foundations, as the most common type of deep foundation, have been applied in geotechnical engineering for ages. Piles are suitable to be used as structural support in frost-susceptible soils in deep seasonally frozen ground or in ice-rich permafrost.Pile foundations can be installed with minimum disturbance to permafrost and can isolate the structure from seasonal frost, subsidence movements of the active layer, and permafrost degradation. It was found that the installation of a pile foundation leads to a reduction in the heaving of frozen soil adjacent to the pile (Shulyat'evet al., 1991). Structural loads are transferred via the pile shaft to the deep layer, where the soil's supporting strength remains stable through the design life of superstructures.

Discussions regarding pile design and standard test procedures in cold regions have been in the forefront since the last century (Nixon and McRoberts,1976; Nixon and Neukirchner, 1984; Bro, 1985;Crory, 1985; Biggar and Sego, 1989; Thomas and Phukan, 1992). In general, two criteria must be satisfied for pile design in permafrost terrain. First, an ad-equate margin of safety is required against gross failure. Second, the settlement must be held within a tolerable limit over the design life of the structure. It was recognized that appropriate design parameters can be determined either from a large number of pile-loading tests or from theoretical solutions if the creep and physical properties of the frozen ground are known.Such factors as temperature profile, external loading,salinity and grain size of soil,etc., can affect a pile-frozen soil system. At the same time, as some facets of frozen soil remain elusive, deeper insights accounting for a combination of various factors are encouraged to quantify pile behaviors. Principles of pile design and engineering practice in cold regions have been summarized, and key parameters clarified(Phukan and Ladanyi, 1992; Cao and Wei, 2004;Scarr and Mokwa, 2008).

The main objective of this paper is to review previous research regarding the behavior and associated engineering practice of pile foundations in cold regions. This paper addresses six facets: (1) bearing capacity and creep, (2) frost jacking, (3) piles under horizontal loads, (4) dynamic properties, (5) hydration heat and refreezing, and (6) countermoves and optimal design. Further studies and applications with respect to piling in cold regions are thus recommended,based upon current practice and issues. Cold regions shall be divided into two fundamental types, namely seasonally frozen regions and permafrost terrain, in terms of the existence of permanently frozen soil beneath the active layer. Such disparity would have critical effects on the serviceability and performance of deep foundations. The physical properties of frozen soil,e.g., modulus, shearing strength, and creep rate,are quite different from those of thawed soil. The stress state at the pile-soil interface is changed when the lower part of a pile is embedded in the frozen layer, and the behavior of the entire pile-soil system is further permanently altered. In the following parts of the paper, sections 2 and 6 mainly deal with practice in permafrost; and section 3 generally describes the engineering hazards that piles suffer from in seasonally frozen regions, while sections 4, 5, and 7 involve both scenarios.

2　Bearing capacity and creep

The bearing capacity of piles embedded in frozen soil is dependent on the long-term adfreeze bond strength at the pile-frozen soil interface and the endbearing capacity. Hence, it can be calculated by suitably combining contributions of these two mechanisms, as determined from short- or long-term creep tests. Researchers generally calculate a pile's bearing capacity by semi-empirical, empirical, numerical, and analytical methods. Several factors such as the configuration and roughness of the pile, negative temperature, ice and mineral volume fractions, and load geometry can affect adfreeze bond strength. Three components contributing to adfreeze bond strength were manifested in a loading test, namely friction, ice cohesion, and mechanical interaction forces arising from surface roughness (Andersland and Alwahhab, 1984).

Loading tests are generally applied to reveal the factors related to bearing capacity of a pile in cold regions. Parameswaran (1979) evaluated the influence of pile materials by a creep loading test, in which model piles made of wood, concrete, and steel were embedded in artificially frozen sand at a temperature of -6 or -10 °C; pile responses and behaviors were monitored and appraised under sustained loads.Moreover, model piles rooted in various types of frozen soils were tested under a constant rate of displacement and load; and an empirical formula was proposed to calculate the ultimate bearing capacity of piles in permafrost (Parameswaran, 1986). Nixon(1988) found that significant saline concentration in the frozen soil, generally exceeding 10-20 ppt, can reduce the bearing capacity and strength of saline permafrost significantly, which was further verified through pile-loading cases. To clarify the influence of pile spacing on bearing capacity, formulae for a single pile and a pile group were proposed separately, based on theoretical investigation (Vyalovet al., 1991). A long-term compression test in saline permafrost was carried out on bored-pipe piles backfilled with diverse materials (Biggaret al., 1996), and the results were compared with those of model pile tests (Biggar and Sego, 1994). Titi and Wathugala (1999) thought that pile setup in cold regions was significant for bearing capacity, which was increased due to soil consolidation for piles driven in saturated clays or silts. Furthermore, a numerical procedure was proposed to predict a pile's bearing capacity during different life stages.

Diversified testing techniques were implemented to clarify bearing capacity and creep behaviors of pile foundations. Isaevet al. (1987) examined by the static-penetration method the bearing capacity of piles embedded in plastic-frozen soil. Similarly, Holubec and Noel (1992) performed load testing on grouted piles, located at ten sites that encompassed various permafrost conditions across the western Canadian Arctic. The results revealed the impact of different test procedures on the assessment of creep rate. In contrast, Vyalov and Mirenburg (1990) suggested an improved testing method that avoids the unpredictably long duration of conventional tests for the determination of bearing capacity of piles in frozen soil.This approach followed the premise that the development of settlement has a certain relation with the decrease in loading, which can be established during the investigation process. A pile load test was conducted to determine the driving depth and bearing capacity of open-ended steel piles in the context of the Kharyaga pipeline terminal (Thompson and Tart, 1996). Given that the oil field was overlain by warm permafrost,piles were adopted to address settlement issues. Similar investigations have also been performed by Merrillet al. (1999), using some advanced measuring techniques known as the pile-driving analyzer (PDA) and the Case Pile Wave Analysis Program (CAPWAP).

The authors believe one of the first comprehensive computational methods was achieved by Wuet al.(2005). In their studies, long-term performance of a reinforced-concrete pile was simulated under prevailing permafrost scenarios, through hydro-thermomechanical (HTM) coupling strategies. A viscoelastic constitutive model was introduced to the frozen volume, and the Goodman element was utilized to describe the interface-shearing behaviors within a pile-soil system. However, this model was simplified based on a partial hypothesis, including insufficiently complete accounting of the frost-heave model and interface theory. It cannot be denied that this work is still a promising step toward dissecting the behavior of a pile subjected to a freezing climate. A preliminary numerical attempt by Wang (2005) and Xu (2006)revealed the stress state of loading a pile embedded in frozen ground. Though a relatively advanced constitutive model for frozen soil was involved, these works lack adaptability, as neither water transport nor thermal effect was considered. A finite-element (FE)program code was developed by Xuet al. (2007), and the results fitted well with field load-displacement curves. Nonlinear relations were introduced to interface elements derived from laboratory adfreeze testing. Jiaet al. (2007) outlined similar work using FE software, employing a built-in interface model. By simulations, Jiang and Guo (2016) predicted the longterm bearing capacity of a cast-in-place pile on the Qinghai-Tibet Plateau; countermeasures are recommended where piles suffer from slumping induced by global warming.

In recent years, some novel investigations were implemented. Shishov (2010) examined the deformation of a compression-bent pile, of which the lower end was rigidly fixed in permanently frozen ground,with the upper end-hinge connected to the rafter of a building; and he presented an analytic solution of the pile's bearing capacity for a given stiffness and strength reserve. Miet al. (2010) conducted experiments on elastic modulus and strength of materials taken from pile foundations in the field; then, the results were employed to aid static-load testing at an outdoor site. Based onin-situdata of high-speed railway(HSR) bridge piles in permafrost, postconstruction settlement was predicted by analyzing settlement data via statistical methods (Lenget al., 2011). Wuet al.(2016) developed a thermal device that can produce heat inside the soil volume. With the device installed beneath the pile toe of a model pile-testing system,the negative impact of thawing soil on bearing capacity of the pile was interpreted under artificial freezing conditions.

In addition to compression, piles loaded in tension showed similar responses, except that toe resistance was not included during this process. Poorooshasb and Parameswaran (1982) appraised the uplift behavior of a rigid pile in frozen sandy soil via theoretical investigation of the displacement field. Accounting for the creep of frozen soil, Sunet al. (2007)presented a nonlinear analysis of uplift piles in permafrost terrain.

Essentially, interface models are crucial to computational approaches to pile foundations. Puswewala and Rajapakse (1990) developed an FE computer code to analyze the interaction between embedded cylindrical foundations and frozen media. The foundation was static-loaded, and displacement-time relationships showed the significant effects of creep parameters on settlement behaviors. Biggar and Kong(2001) presented the results of long-term load tests during the installation of short-range radar facilities across the Canadian Arctic, and the results were then compared to design guidelines for ice-rich, saline permafrost. A multitude of scholars (Jianet al., 2002; Luet al., 2007; Yueet al., 2015) employed various methods to examine shearing and sliding over the pile surface, though few could be associated with freezing scenarios.

To dissect the bearing capacity of piles embedded in frozen ground, reliable data collected fromin-situmonitoring or laboratory testing are required, which can also help quantify parameters adopted in numerical approaches. The frozen soil-pile interface is not fully understood, and the inherent mechanism calls for a more sophisticated theoretical explanation. For the interface theory, physicochemical processes can be involved at freezing.

Existing experimental investigations of the interface's shearing characteristics (Wenet al., 2013; Sunet al., 2014; Lü and Liu, 2015; Chenet al., 2016; Shiet al., 2016; Jiet al., 2017) were centered on the direct-shearing test having a shear rate greater than 0.02 mm/min, which cannot reflect the real state the pile surface undergoes in engineering practice. The creep behaviors of piles subjected to sustained loads cannot be linked to direct-shear tests on frozen soil-pile interface under a constant strain rate (Heet al., 2016).As a result, the time factor should be considered in the experiments and interface model before adopting it for further determination of the bearing capacity of piles in frozen soil. It is also suggested that the role of ice in the adfreeze bond should be evaluated separately. The effects of temperature and time on icebonding need to be precisely quantified, apart from the mechanical interlocking and cohesion on the shearing surface.

3　Frost-jacking phenomenon

During winter, when cooling temperature penetrates into frost-susceptible soil, significant frost heave can be anticipated if sufficient water is available.Consequentially, a deep foundation can be ejected from the frozen ground due to upward movement of the surrounding soil. This phenomenon is generally referred as frost jacking, which can be attributed to the basal freezing pressure applied at the bottom of the foundation, or to the tangential frost-heave force acting on the side of a foundation. As the temperature decreases, a strong adfreeze bond develops between frozen soil and the pile, giving rise to a dramatic increase in the pile's bearing capacity, as well as frostjacking hazards, which should not be negligible.

To address the abovementioned challenges frequently encountered in seasonally frozen regions,many investigations have been made, including experimental field studies and positing of potential mechanisms (Dalmatovet al., 1973; Domaschuk, 1982;Fukuda and Kinosita, 1985). Their findings demonstrated that the main factors affecting adfreeze shear stress within the freezing zone were (1) soil temperature; (2) displacement rate at the pile-soil interface; (3)interface characteristics; and (4) slip or nonslip conditions, depending on the amount of displacement. Recent research on shear strength of the structure-frozen ground interface found that it is also strongly influenced by the temperature and moisture content of frozen soil. Variations in the shear strength with temperature and moisture content mainly resulted from the variation in the cohesive component of the ice(Wenet al., 2016). Field observations deduced that the maximum values of the mobilized adfreeze shear stress occurred at the initial stage, with a high heaving rate; the peak uplift forces were reached when frost penetration was close to the maximum, having a low heaving rate.

In essence, frost uplift responds in a strong,though hysteretic, relation to ground temperature variations (Crory and Reed, 1965); heave force continues to grow even after the frost penetration has attained the permafrost table. Accounting for the size effect,the mobilized adfreeze-strength magnitude is inversely proportional to the pile diameter (Penner,1974). Some valuable attempts have been made by several scholars (Penner and Gold, 1971; Fukuda and Kinosita, 1985; Johnson and Esch, 1985) to simulate the development of frost-heave forces on piles. Ladanyi and Foriero (1998) proposed a closed-formed solution to calculate the development of adfreeze frost-heaving force operating on a fixed pile, accounting for the effects of heave rate, frost-penetration rate,and soil temperature at any depth along the shaft.Ashkinadze (2011) investigated the performance of a friction pile in clayey soil under the combined action of frost jacking and transient loads, by introducing an analytical model named the "three-column model".Three limit states were considered for pile-design procedure and risk assessment. To assess the risk of frost jacking and provide reasonable suggestions for design and construction, a safety-coefficient calculation that involves frost penetration, frost-heave force, and freezing strength, as well as loads, was proposed(Wenet al., 2012). Countermeasures to mitigate the uplift of piles subjected to frost jacking include applications of belled piles (Segoet al., 2003) and screw piles (Wanget al., 2016a) instead of conventional straight-shafted piles. In seasonally frozen regions,these piles having specific configurations can be anchored into unfrozen layers beneath the maximum depth of frost penetration, providing substantial resistance against the uplift force induced by frost heave.

Shearing behavior at the interface when a pile is jacked up due to frost heave is elusive for two scenarios, either where the uplift displacement is small or where it is too large. In the latter case, the determination of the limit value of shear strain, when the corresponding stress attains the peak value, is of significant importance. Previous studies suggest that the adfreeze bond after bond breaking was considerably low(Nidowicz and Shur, 1998). However, the test results showed that a signi fi cant bond recovery was observed; and the residual adfreeze stress was still very high after the initial bond was broken, especially for frozen soils having a high moisture content (Wenet al., 2016). When the environmental temperature is low, and the displacement rate is nominal, the melting of ice and the rupture of ice bonding are accompanied by new formation of ice crystals during the shearing process. So special attention should be paid to the phenomenon of bond healing, in terms of pile design in frozen ground (Ladanyi and Theriault,1990). In addition, residual bond strength after sliding should be quantified for diverse soil types in practice.

4　Laterally loaded piles

Piles subjected to lateral loads in cold regions are of particular interest in association with aboveground oil pipelines, heated structures supported by piles, and power poles, as well as offshore drilling platforms.Horizontal loads arising from wind, ice, vehicular impact, waves,etc., are generally the critical factor for the design of such structures.

The performance predictions of laterally loaded piles are generally derived from two prevailing methods adopted for piles embedded in thawed/unfrozen soils, namely modulus of subgrade and pressure deflection-curve methods. Loading tests were carried out to determine creep parameters for pipe piles; then,the loading capacity of lateral piles was predicted based on the acquired parameters (Rowleyet al.,1975). Another approach for determining the creep parameters came from pressuremeter creep tests. A reasonable agreement was found between these two methods. Ladanyi (1985) discussed the possibility of using CPT (cone penetration test) in connection with the design of lateral piles.

With a computer program that employedp-ycurves to describe the loading process and pile-soil interactions, Crowther (1990, 2015) presented a design approach to analyze the effects of horizontal loads on piles in frozen soils. Shear strength and strain criteria were introduced to constructp-ycurves and were parameterized based on thermal regime and load duration. Ap-ymodel was developed by Shelmanet al. (2014), based on the constitutive relationship of frozen soils; then, it was applied to assess the seismic performance of bridge L-pile foundations, accounting for variations of soil temperature. Similar thoughts on the application of thep-yapproach in analysis of laterally loaded piles in frozen silt have been shared by Li and Yang (2017). Furthermore,Yanget al. (2017) carried out a large pile-deformation test to gatherin-situdata regarding lateral performance of piles under frozen and unfrozen conditions, respectively, and demonstrated how the collected data can be used in a simplified design tool.

By contrast, the numerical solutions for piles subjected to lateral loads in permafrost developed quickly, usually achieved by means of finite-difference techniques applied to solving coupled partial differential equations. A preliminary finite-element method (FEM) study concerning creep behaviors was presented by Foriero and Ladanyi (1990), using a Maxwell model, which was then verified by results of loading cases. Subsequently, Foriero and Ladanyi(1991a) tried to propose a general procedure for lateral-stability analysis of a single pile subjected to both axial and horizontal loads, whereas such quantities as equivalent-length factor and buckling length can be easily calculated to facilitate the analysis using design code for the practice. Later, an FE model to simulate the behavior of laterally loaded piles in permafrost was developed, for the specific purpose of verifying an analytical method derived from streamline theory(Foriero and Ladanyi, 1995). The results indicated that the assessment of the interface adfreeze and friction component required further research. In northern Quebec, a field study was undertaken on laterally loaded piles in permafrost, under a controlled displacement rate (Forieroet al., 2005). The test results,including bending moments, shear loads, and pile displacements, were employed to verify an earlier FEM model developed by Foriero and Ladanyi (1991b). Guet al. (2016) presented an FE simulation to appraise nonlinear behaviors of a steel-pipe pile tested at a field site in Alaska. Sensitivity-analysis methodology was adopted to delve into frozen soil-pile interactions when piles were subjected to lateral loads.

In many cases, piles suffer from lateral loads and axial loads simultaneously, especially when an earthquake takes place. We should make an effort to develop comprehensive approaches, thus such conditions can be adequately appraised, and the contributions of horizontal and vertical loads clarified.

5　Dynamic response of pile-frozen soil systems

In Arctic and subarctic regions covered by seasonally frozen ground or permafrost, some structural failures during earthquakes appear to be directly attributed to the existence of frozen ground and ice formation. Because research has documented the significant effect of frozen soil on the seismic response of bridge foundations, it is vital to evaluate the dynamic performance of piles under frozen scenarios. The reason lies in that the ice matrix formed within the soil volume during freezing greatly affects soil-structure interactions (SSI) due to higher stiffness and shear strength with respect to unfrozen soil.

Parameswaran (1984) conducted a creep test of piles in frozen soil subjected to static loads and superimposed alternating loads, the results of which indicated that a dynamic stress as small as 5% of the staticstress magnitude accelerates the displacement and creep rate, thereafter decreasing the long-term bearing capacity of piles in permafrost. A similar experiment was conducted, only differing in pile materials and the amplitude of cyclic loads (Parameswaran,1982). Later, Stelzer and Andersland (1991) further revealed the sliding mechanism operating at the pile-frozen sand interface by applying both static and superimposed cyclic loads to friction piles. It indicated that the addition of cyclic loads would greatly increase the displacement rate over those observed under sustained loads.

One of the first dynamic tests on driven piles in permafrost was conducted by Merrill and Riker(1996). It allowed assessments of pile integrity during installation, thus serving as a useful tool to prevent possible damage to piles under permafrost conditions. Zhanget al. (1999) performed model pile tests to investigate several factors that might contribute to the dynamic response of piles under cyclic loads, in-dicating that the creep rate of model piles was independent of the frequency of alternating loads. Wuet al. (2010b) conducted scale shaking table tests for model piles under negative temperature, and found that seismic load accounted for an increase in temperature at the pile-soil interface; and warm permafrost was quite sensitive to seismic motions, which implied potential instability. Chenet al. (2012) carried out a free-vibration pile test, having 80% of the lower portion of the pile embedded in frozen Fairbanks silt. The dominant system parameters that characterize pile-frozen soil interactions were identified by a freedecay response-signal approach. Liet al. (2012) developed a model test system to evaluate dynamic behavior of the pile-frozen soil interface, and it could be extended to other practices in cold regions.

It has been demonstrated through analytical and experimental studies that the dynamic response of bridge columns supported by pile foundations is affected by seasonal frost. Therefore, it is crucial to dissect the interaction mechanism and seismic performance of a soil-foundation-structure (SFS) system. A selected bridge located in Alaska was continuously monitored to explore such environmental variables as traffic-induced vibrations, ambient noise, and earthquakes on dynamic performance of the SFS system(Yanget al., 2007). Then, an elastic-plastic FE analysis was carried out to obtain simulation results from a soil-pile system under the same condition (Xiong and Yang, 2008), the results of which compared favorably with those obtained from the field. A numerical model was developed to assess the lateral cyclic behaviors of a simple two-span prototype bridge system under frozen and unfrozen conditions, respectively (Wotherspoonet al., 2010). The results showed that the frozen condition increased the peak bending moment and shear-stress demands for all seismic intensities, compared with the unfrozen condition. Similar scenarios have been worked on by Yanget al.(2014) through fluid-solid coupled FE analysis. Shelmanet al. (2014) examined the monotonic characteristics of several soil types with a function of negative temperature, so that a suitablep-ycurve can be developed for design guidelines.

The dynamic properties of a frozen soil-pile system can be sufficiently clarified by introducing the freezing effects into a common pile-soil system, with dominant factors highlighted. Numerical simulations should emphasize dynamic shearing behavior at the interface.

6　Impact of hydration heat and the refreezing process

Due to the continuous demand for economic development and natural resources, construction activities of transportation have seen rapid growth in cold regions (Shanget al., 2017). It was reported that castin-place (CIP) piles accounted for 90% of bridge foundations on the Qinghai-Tibet Plateau (Wanget al., 2003). In the majority of permafrost regions of the world except China, precast pile, primarily made of timber or steel, is the most common type of foundation. Hence, research on CIP piles was usually carried out by Chinese scholars. It is observed that the construction of pile foundations for bridges disturbs adjacent soils in permafrost terrain, arising from pile installation and concrete-hydration heat. As the surrounding soil melts due to rising temperature, adfreeze bond strength might not be sufficient, resulting in lower bearing capacity of pile foundations. It follows that settlement or tilt of the foundation would take place if no proper measures are taken during or after construction. Consequently, a host of researchers and engineers tried to find feasible solutions to minimize the elevated temperature of frozen soil induced by hydration heat and drilling (Liuet al.,2015).

Yuet al. (2007) and Wanget al. (2013) performed field tests to evaluate the bearing capacity of a large-scale CIP bored pile, and dissected shaft resistance and tip resistance when adjacent soils were not fully refrozen soon after construction. Some researchers, including Wuet al. (2004, 2010a), Xionget al.(2009), and Liu (2013), employed numerical methods to analyze the temperature regime of a pile-soil system affected by hydration heat and the refreezing process on the Qinghai-Tibet Plateau. Latent heat and energy released by the concrete-hydration process were considered for the calculation of temperature distributions radially or along the depth. Chenet al.(2014) evaluated impacts of such factors as molding temperature, hydration heat, and volumetric ice content on thawing depth and temperature regime of pile foundations during construction. To predict the refreezing time and ultimate bearing capacity of bored piles for a bridge rooted in polygonal permafrost regions, static- and dynamic-load tests were undertaken for full-scale test piles (Yuet al., 2015). Fuet al.(2016) investigated the impact of hydration heat and the refreezing law for a large-scale CIP pile by a comparison ofin-situmeasurements and numerical simulations.

A breakthrough in this aspect generally requires proper installation techniques and products of highquality concrete. The primary principle is to preserve the thermal state of permafrost during and after the installation of CIP piles.

7　Optimal design and countermeasures

Pile foundations in cold regions are facing di-verse challenges, which can mainly be attributed to frost heave, thaw settlement, effects of freeze-thaw cycles, and creep/degradation of the permafrost. In general, measures to deal with issues occasionally encountered before or after operation can be divided into two types: one is associated with the optimized design of piles; and the other is connected to adjacent soils, usually in the form of frozen ground.

In addition to conventional piles adopted in engineering practice, novel pile configurations and geomaterials have been developed to enhance pile performance. Fedorvichet al. (1987) proposed one specific pile having a special configuration to satisfy requirements based on extreme conditions in the northern part of west Siberia. Bastidas (1989) worked on a case history in which an innovative pile-foundation design approach was raised up, applied in the Prudhoe Bay Field, to ideally reverse the warming trend of the subsoil, which imposed a severe threat to the long-term bearing capacity of piles. Novel geomaterials have been introduced to concrete-composite piles against complex environments,e.g., a combination of constant loads and freeze-thaw cycles. This novel geomaterial, known as concrete-filled, fiber-reinforced-polymer tubes (CFFTs), was proved effective by experimental investigations (Famet al., 2008). The experience obtained from the Qinghai-Tibetan grid project indicated that a fi berglass-reinforced plastic cover can signi fi cantly reduce the adfreeze bond strength and the tangential frost-heave force due to the adfreeze bond (Wenet al., 2016). A novel type of carved, grooved pile made of concrete is refilled with coarse-grained porous materials to help mitigate degradation of the permafrost and downdrag force induced by thaw settlement of subsoil on the Qinghai-Tibet Plateau (Li and Xu, 2008). Belled piles are effective to inhibit upward movements caused by the frost-jacking process (Segoet al., 2003), as the special structure with an enlarged base counteracts uplift force due to frost heave of soil. A similar principle is also valid for the helical pier, which has been investigated experimentally (Wanget al., 2017a) and numerically (Wanget al., 2017b). Helical piers are widely adopted in seasonally frozen ground to support lightweight superstructures.

Another feasible approach is to preserve the frozen ground or to replace it with non-susceptible soil. Song (2014) presented several anti-freezing measures in seasonally frozen soils, namely geotextile-pile wrapped with residue oil, and piles filled with ash and Aeolian sand, respectively. The performances of the countermeasures were compared to determine the most effective approach. Thermosyphon is regarded as impactful tool to cool subsoil around the foundations in permafrost. It is usually a sealed tube having a circulation of working fluid inside.Field observations of the soils close to the footings along the Qinghai-Tibet Power Transmission Line showed reasonable cooling effects, which helped to maintain the frozen state of the sub-base soil (Guoet al., 2016). Muet al. (2016a) presented numerical simulations to study the cooling effect of thermosyphons and their application in perennially frozen regions.Further insights were gained by investigating a combined cooling strategy of thermosyphons and insulation boards (Muet al., 2016b). Perreault and Shur(2016) suggested using seasonal thermal insulation to mitigate temperature-variation impacts on foundations in cold regions. Wanget al. (2016b) carried out field tests to study the effects of changing backfills,which benefited the stability of CIP bulb piles embedded in ice-rich frozen soil.

Novel geomaterials like geosynthetics are strongly recommended to mitigate frost-related issues, as they are usually cost-effective. For some cases, a combined method is more reasonable than a separate treatment. The authors would like to propose the idea of a"smart pile" that essentially combines a conventional pile with a thermosyphon. It is expected to provide bearing capacity, as well as preserving the thermal state of the surrounding frozen soil proactively through the seasons.

8　Concluding remarks

From the review of research on pile applications in cold regions, it can be summarized that the developments with respect to piling are usually accompanied by advances in techniques and emerging issues arising from engineering practice. Advanced monitoring technologies and the aid of computing techniques enable us to analyze pile performance in seasonally frozen/permafrost regions in a more effective way.The hazards induced by concrete-hydration heat occurring after CIP piles have prevailed, whereas optimal design and countermeasures are desirable to mitigate such an impact. Suggestions regarding future work are as follow:

1) Advanced and reliable measuring methods should be developed for assessing the long-term bearing capacity of piles. Connections should be established between short- and long-term loading tests. A sophisticated interface model to describe frozen soil-pile shearing behaviors under different scenarios is an urgent need, as all numerical approaches rely heavily upon it. One feasible way is to examine the mechanism at a microlevel, especially microscopic structure change and its mechanism at the interface between the pile and frozen soil subjected to longterm loads.

2) A more comprehensive frost-heave model of soil is an urgent need when simulating the frost-jack-ing process, along with a constitutive relationship to describe the creep behavior.

3) A smart pile needs to be invented. It is recommended to design one kind of pile that possesses the basic bearing capacity, as well as the ability to convert the thermal regime of adjacent soil into a more favorable condition.

This work was supported by the National Natural Science Foundation of China (Grant Nos. 41731281 and 41771073).

Andersland OB, Alwahhab MRM, 1984. Load capacity of model piles in frozen ground. In: Proceedings of 3rd International Specialty Conference. Canadian Society for Civil Engineering, pp. 29-39.

Ashkinadze K, 2011. Limit states design of pile foundations in clay under combined action of transient uplift loads and frost jacking. In:Geo-Risk 2011: Risk Assessment and Management. Atlanta Georgia, United States: ASCE, pp. 656-663.

Bastidas CP, 1989. Innovative foundation designs for permafrost conditions in the arctic. Foundation Engineering: Current Principles and Practices, 2: 1643-1657.

Biggar KW, Sego DC, 1989. Field load testing of various pile configurations in saline permafrost and seasonally frozen rock. In: Proceedings of the 42nd Canadian Geotechnical Conference, Canadian Geotechnical Society, pp. 304-312.

Biggar KW, Sego DC, 1994. Time-dependent displacement behaviour of model adfreeze and grouted piles in saline frozen soils. Canadian Geotechnical Journal, 31(3): 395-406. DOI: 10.1139/t94-047.

Biggar KW, Sego DC, Stahl RP, 1996. Long-term pile load testing system performance in saline and ice-rich permafrost. Journal of Cold Regions Engineering, 10(3): 149-162.

Biggar KW, Kong V, 2001. An analysis of long-term pile load tests in permafrost from the Short Range Radar site foundations. Canadian Geotechnical Journal, 38(3): 441-460. DOI: 10.1139/t00-110.

Bro A, 1985. Research challenges in frozen ground pile design. In:Proceedings of Civil Engineering in the Arctic Offshore. ASCE,pp. 1196-1205.

Cao YX, Wei QC, 2004. Study on design of pile foundation in permafrost area on Qinghai-Tibet railway line. Journal of Railway Engineering Society, 21(2): 127-131. DOI: 10.3969/j.issn.1006-2106.2004.02.030.

Chen GS, Davis D, Hulsey JL, 2012. Measurement of frozen soil-pile dynamic properties: a system identification approach. Cold Regions Science and Technology, 70: 98-106. DOI: 10.1016/j.coldregions.2011.08.007.

Chen LZ, Wen Z, Dong SS,et al., 2016. Study of the constitutive model of interface between frozen Qinghai-Tibetan silt and reinforced plastic fiberglass. Journal of Glaciology and Geocryology, 38(2):402-408. DOI: 10.7522/j.issn.1000-0240.2016.0044..

Chen ZY, Li GY, Mu YH,et al., 2014. Impact of molding temperature and hydration heat of concrete on thermal properties of pile foundation in permafrost regions along the Qinghai-Tibet DC transmission line. Journal of Glaciology and Geocryology, 36(4): 818-827.DOI: 10.7522/j.issn.1000-0240.2014.0098.

Crory FE, Reed RE, 1965. Measurement of Frost Heaving Forces on Piles. CRREL Report, pp. 145.

Crory FE, 1985. Standards for pile load tests in permafrost. Foundations in Permafrost and Seasonal Frost. ASCE, pp. 15.

Crowther GS, 1990. Analysis of laterally loaded piles embedded in layered frozen soil. Journal of Geotechnical Engineering, 116(7):1137-1152.

Crowther GS, 2015. Lateral pile analysis frozen soil strength criteria.Journal of Cold Regions Engineering, 29(2): 04014011.

Dalmatov BI, Karlov VD, Turenko II,et al., 1973. Interaction of freezing heaving with foundations. In: Proceedings of the 2nd International Conference on Permafrost Yakutsk, USSR, North American Contribution. Washington, DC: National Academy of Science, pp.572-576.

Domaschuk L, 1982. Frost heave forces on embedded structural units.In: Proceedings of the 4th Canadian Permafrost Conference. Calgary, Alta: National Research Council of Canada, pp. 489-496.

Fam A, Kong A, Green, MF, 2008. Effects of freezing and thawing cycles and sustained loading on compressive strength of precast concrete composite piles. PCI Journal, 53(1): 109-120.

Fedorovich DI, Targulyan YO, Savel'ev VS,et al., 1987. Improved types of piles for construction on permafrost soils in the north of West Siberia. Soil Mechanics and Foundation Engineering, 24(5):171-176. DOI: 10.1007/BF01787137.

Foriero A, Ladanyi B, 1990. Finite element simulation of behavior of laterally loaded piles in permafrost. Journal of Geotechnical Engineering, 116(2): 266-284.

Foriero A, Ladanyi B, 1991a. Design of piles in permafrost under combined lateral and axial load. Journal of Cold Regions Engineering,5(3): 89-105.

Foriero A, Ladanyi B, 1991b. Generalized finite-element algorithm for laterally loaded piles in permafrost in comparison with measurements. Canadian Geotechnical Journal, 28(4): 523-541. DOI:10.1139/t91-069.

Foriero A, Ladanyi B, 1995. FEM simulation of interface problem for laterally loaded piles in permafrost. Cold Regions Science and Technology, 23(2): 121-136. DOI: 10.1016/0165-232X(94)00008-L.

Foriero A, St-Laurent N, Ladanyi B, 2005. Laterally loaded pile study in permafrost of Northern Québec, Canada. Journal of Cold Regions Engineering, 19(3): 61-84.

Fu J, Jiang Y, Peng H,et al., 2016. Early refreezing law of large-diameter cast-in-place piles in permafrost regions. Journal of Traffic and Transportation Engineering, 16(4): 104-111.

Fukuda M, Kinosita S, 1985. Field prediction of the uplift force to conduits due to frost heaving. In: Proceedings of the 5th International Symposium on Ground Freezing. Sapporo, Japan, pp. 135-139.

Gu Q, Yang ZH, Peng Y, 2016. Parameters affecting laterally loaded piles in frozen soils by an efficient sensitivity analysis method.Cold Regions Science and Technology, 121: 42-51. DOI:10.1016/j.coldregions.2015.10.006.

Guo L, Yu QH, You YH,et al., 2016. Cooling effects of thermosyphons in tower foundation soils in permafrost regions along the Qinghai-Tibet Power Transmission Line from Golmud, Qinghai Province to Lhasa, Tibet Autonomous Region, China. Cold Regions Science and Technology, 121: 196-204. DOI:10.1016/j.coldregions.2015.07.011.

He F, Wang X, Jiang DJ,et al., 2016. Thoughts on the research of behaviors of frozen soil-structure interface. Chinese Journal of Underground Space and Engineering, 12(S1): 133-139.

Holubec I, Noel M, 1992. Pile load testing of grouted pile across western Canadian Arctic. In: Proceedings of the Canadian Geotechnical Conference. 15: 1-10.

Isaev ON, Volkov FE, Minkin MA, 1987. Static-penetration determination of the bearing capacity of piles in plastic-frozen soils. Soil Mechanics and Foundation Engineering, 24(5): 194-198. DOI:10.1007/BF01787142.

Ji YJ, Jia K, Yu QH,et al., 2017. Direct shear tests of freezing strength at the interface between cast-in-situ concrete and frozen soil. Journal of Glaciology and Geocryology, 39(1): 86-91. DOI:10.7522/j.issn.1000-0240.2017.0011.

Jia YM, Guo HY, Guo QC, 2007. Finite element analysis of bored pile-frozen soil interactions in permafrost. Chinese Journal of Rock Mechanics and Engineering, 26(S1): 3134-3140. DOI:10.3321/j.issn:1000-6915.2007.z1.082.

Jian Z, Liu YN, Li JH, 2002. Pile-soil 3-D constant thickness contact interface element model and its characteristics. Journal of Hydraulic Engineering, 33(11): 113-118. DOI: 10.3321/j.issn:0559-9350.2002.11.021.

Jiang DJ, Guo CX, 2016. Long-term stability of bearing capacity of single pile in permafrost area on Qinghai-Tibetan Plateau. Journal of Chang'an University (Natural Science Edition), 36(2): 59-65.

Johnson JB, Esch DC, 1985. Frost jacking forces on H and pipe piles embedded in Fairbanks silt. In Proceedings 4th Symposium on Ground Freezing, Sapporo. V.2, pp. 125-133.

Ladanyi B, 1985. Use of the cone penetration test for the design of piles in permafrost. Journal of Energy Resources Technology,107(2): 183-187. DOI: 10.1115/1.3231174.

Ladanyi B, Foriero A, 1998. Evolution of frost heaving stresses acting on a pile. In: Proceedings of the PERMAFROST-Seventh International Conference. Yellowknife, Canada: Collection Nordicana, 55:623-633.

Ladanyi B, Theriault A, 1990. A study of some factors affecting the adfreeze bond of piles in permafrost. In: Proceedings of the Geotechnical Engineering Congress. GSP 27 ASCE, 327-334.

Leng WM, Yang Q, Nie RS,et al., 2011. Study of post-construction settlement combination forecast method of high-speed railway bridge pile foundation. Rock and Soil Mechanics, 32(11):3341-3348. DOI: 10.3969/j.issn.1000-7598.2011.11.023.

Li N, Xu B, 2008. A new type of pile used in frozen soil foundation.Cold Regions Science and Technology, 53(3): 355-368. DOI:10.1016/j.coldregions.2007.10.005.

Li Q, Yang ZH, 2017. P-Y approach for laterally loaded piles in frozen silt. Journal of Geotechnical and Geoenvironmental Engineering,143(5): 04017001. DOI: 10.1061/(ASCE)GT.1943-5606.0001556.

Li YB, Zhang HR, Quan KJ, 2012. Development of model test system for dynamic frozen soil-pile interaction. Chinese Journal of Geotechnical Engineering, 34(4): 774-780.

Liu L, 2013. Study on stability for saline soil subgrade and pile foundation in permafrost of cold and arid region. Lanzhou: Lanzhou Jiaotong University, pp. 19-43.

Liu WB, Yu WB, Yi X,et al., 2015. Thermal regime of frozen soil foundation affected by concrete base of transmission line tower on the Tibetan Plateau. Applied Thermal Engineering, 75: 950-957.DOI: 10.1016/j.applthermaleng.2014.10.041.

Lu HX, Shang SP, Yu J, 2007. Research on planar contact model of dynamic soil-pile interaction. Engineering Mechanics, 24(9):129-133. DOI: 10.3969/j.issn.1000-4750.2007.09.020..

Lü P, Liu JK, 2015. An experimental study on direct shear tests of frozen soil-concrete interface. Journal of the China Railway Society, 37(2): 106-110. DOI: 10.3969/j.issn.1001-8360.2015.02.20..

Merrill KS, Riker RE, 1996. Comparison of static and dynamic test results for driven steel pipe piles in highly saline permafrost. In:Proceedings of the 8th International Conference on Cold Regions Engineering. Fairbanks, Alaska: University of Alaska, Fairbanks,pp. 238-253.

Merrill KS, Korri K, Miner RF, 1999. Installation and evaluation of driven steel pipe piles in Alaska soils. In: Proceedings of the International Conference on Cold Regions Engineering: Putting Research into Practice. ASCE, pp. 717-730.

Mi WJ, Li Y, Shi GQ,et al., 2010. Study on stability of bridge pile foundation in permafrost area along Qinghai-Tibet Railway. Journal of Railway Engineering Society, 27(9): 15-19. DOI:10.3969/j.issn.1006-2106.2010.09.004..

Mu YH, Yu QH, Ma W,et al., 2016a. Numerical study of long-term cooling effects of thermosyphons around tower footings in permafrost regions along the Qinghai-Tibet Power Transmission Line.Cold Regions Science and Technology, 121: 237-249. DOI:10.1016/j.coldregions.2015.06.005.

Mu YH, Wang GS, Yu QH,et al., 2016b. Thermal performance of a combined cooling method of thermosyphons and insulation boards for tower foundation soils along the Qinghai-Tibet Power Transmission Line. Cold Regions Science and Technology, 121:226-236. DOI: 10.1016/j.coldregions.2015.06.006.

Nidowicz B, Shur YL, 1998. Russia and north American approaches of pile design in relation to frost action. In: Proceeding of the PERMAFROST-Seventh International Conference. Yellownife,Canada: Collection Nordicana, pp. 803-809.

Nixon JF, McRoberts EC, 1976. A design approach for pile foundations in permafrost. Canadian Geotechnical Journal, 13(1): 40-57.DOI: 10.1139/t76-005.

Nixon JF, Neukirchner RJ, 1984. Design of vertical and laterally loaded piles in saline permafrost. In: Proceedings of 3rd International Specialty Conference-Cold Regions Engineering, Northern Resource Development. Canadian Society for Civil Engineering,pp. 131-144.

Nixon JF, 1988. Pile load tests in saline permafrost at Clyde River,Northwest Territories. Canadian Geotechnical Journal, 25(1):24-32. DOI: 10.1139/t88-003.

Parameswaran VR, 1979. Creep of model piles in frozen soil. Canadian Geotechnical Journal, 16(1): 69-77. DOI: 10.1139/t79-007.

Parameswaran VR, 1982. Displacement of piles under dynamic loads in frozen soils. In: Proceedings of the 4th Canadian Permafrost Conference. Calgary, Alta: National Research Council of Canada,pp. 555-559.

Parameswaran VR, 1984. Effect of dynamic loads on piles in frozen soils. In: Proceedings of 3rd International Specialty Conference Cold Regions Engineering, Canadian Society for Civil Engineering. pp. 41-52.

Parameswaran VR, 1986. Bearing capacity calculations for piles in permafrost. In: Proceedings of the 4th International Conference on Cold Regions Engineering. Anchorage, AK, USA: ASCE, pp.751-759.

Penner E, Gold LW, 1971. Transfer of heaving forces by adfreezing to columns and foundation walls in frost-susceptible soils. Canadian Geotechnical Journal, 8(4): 514-526. DOI: 10.1139/t71-053.

Penner E, 1974. Uplift forces on foundations in frost heaving soils. Canadian Geotechnical Journal, 11(3): 323-338. DOI: 10.1139/t74-034.

Perreault P, Shur Y, 2016. Seasonal thermal insulation to mitigate climate change impacts on foundations in permafrost regions. Cold Regions Science and Technology, 132: 7-18. DOI:10.1016/j.coldregions.2016.09.008.

Phukan A, Ladanyi B, 1992. Arctic foundations. Pile foundations in permafrost. In: Proceedings of the 11th International Conference on Offshore Mechanics and Arctic Engineering, ASME. V.4, pp. 125.

Poorooshasb HB, Parameswaran VR, 1982. Uplift of rigid vertical piles in frozen sand. Soils and Foundations, 22(2): 82-88. DOI:10.3208/sandf1972.22.2_82.

Puswewala UGA, Rajapakse RKND, 1990. Numerical modeling of structure-frozen soil/ice interaction. Journal of Cold Regions Engineering, 4(3): 133-151. DOI: 10.1061/(ASCE)0887-381X(1990)4:3(133).

Rowley RK, Watson GH, Ladanyi B, 1975. Prediction of pile performance in permafrost under lateral load. Canadian Geotechnical Journal, 12(4): 510-523. DOI: 10.1139/t75-057.

Scarr KL, Mokwa RL, 2008. Axial capacity of piles founded in permaforst: a case study on the applicability of modern pile design in remote Mongolia. In: Proceedings of the 41st Symposium on Engineering Geology and Geotechnical Engineering. Boise, Idaho, USA,pp. 51-66.

Sego DC, Biggar KW, Wong G, 2003. Enlarged base (belle) piles for use in ice or ice-rich permafrost. Journal of Cold Regions Engineering, 17(2): 68-88. DOI: 10.1061/(ASCE)0887-381X(2003)17:2(68).

Shang YH, Niu FJ, Liu MH,et al., 2017. Long-term effect of a pile foundation on ground temperatures in permafrost regions. Chinese Journal of Rock Mechanics and Engineering, 36(9): 2313-2323.DOI: 10.13722/j.cnki.jrme.2016.1551.

Shelman A, Tantalla J, Sritharan S,et al., 2014. Characterization of seasonally frozen soils for seismic design of foundations. Journal of Geotechnical and Geoenvironmental Engineering, 140(7):04014031. DOI: 10.1061/(ASCE)GT.1943-5606.0001065.

Shi QB, Yang P, Wang GL, 2016. Experimental study on adfreezing strength of the interface between artificial frozen sand and structure. Chinese Journal of Rock Mechanics and Engineering, 35(10):2142-2151. DOI: 10.13722/j.cnki.jrme.2015.1511.

Shishov II, 2010. Bearing capacity of compression-bent piles in permafrost. Soil Mechanics and Foundation Engineering, 47(4): 130-135.DOI: 10.1007/s11204-010-9100-x.

Shulyat'ev OA, Kuzevanov VV, Kemerov VD, 1991. Application of pile foundations with grillage in frost zone of heaving soils. Soil Mechanics and Foundation Engineering, 28(2): 59-62. DOI:10.1007/BF02316014.

Song JQ, 2014. Pile foundation antifreezing measures research in seasonal frozen soil region. Applied Mechanics and Materials, 578-579: 767-771.

Stelzer DL, Andersland OB, 1991. Model pile-settlement behavior in frozen sand. Journal of Cold Regions Engineering, 5(1): 1-13.DOI: 10.1061/(ASCE)0887-381X(1991)5:1(1).

Sun HC, Yang P, Wang GL, 2014. Development of mechanical experimental system for interface layer between frozen soil and structure and its application. Rock and Soil Mechanics, 35(12): 3636-3641.DOI: 10.16285/j.rsm.2014.12.036.

Sun XX, Zhang H, Tian M, 2007. Nonlinear finite element analysis of uplift pile with consideration of creep characteristic in permafrost.Chinese Journal of Rock Mechanics and Engineering, 26(S1):3101-3106. DOI: 10.3321/j.issn:1000-6915.2007.z1.077.

Thomas H, Phukan A, 1992. Arctic foundations. Case studies of pile foundations in permafrost. In: Proceedings of the 11th International Conference on Offshore Mechanics and Arctic Engineering,ASME. V.4, pp. 129.

Thompson SR, Tart RG, 1996. Driven pile capacities in warm permafrost in Komi republic, Russia. In: Proceedings of the 8th International Conference on Cold Regions Engineering. Alaska, USA:ASCE, pp. 254-265.

Titi H, Wathugala G, 1999. Numerical procedure for predicting pile capacity-setup/freeze. Transportation Research Record: Journal of the Transportation Research Board, 1663: 25-32. DOI: 10.3141/1663-04.

Vyalov SS, Mirenburg YS, 1990. Improved methods of testing piles in frozen soils. Soil Mechanics and Foundation Engineering, 27(4):155-161. DOI: 10.1007/BF02305647.

Vyalov SS, Slepak ME, Lunev MV, 1991. Single pile and pile group in permafrost. In: Proceedings of the Sixth International Conference on Cold Regions Engineering. New York: ASCE, pp. 44-53.

Wang JF, Jia K, Rafique R,et al., 2016b. Changes of backfill soil of tower foundation in the permafrost regions with warm ice-rich frozen soil on the Qinghai-Tibet Plateau. Environmental Earth Sciences, 75(21): 1416. DOI: 10.1007/s12665-016-6223-z.

Wang TF, Liu JK, Zhao HG,et al., 2016a. Experimental study on the anti-jacking-up performance of a screw pile for photovoltaic stents in a seasonal frozen region. Journal of Zhejiang University-Science A, 17(7): 512-524. DOI: 10.1631/jzus.A1600407.

Wang TF, Liu JK, Tian YD,et al., 2017a. Frost jacking characteristics of screw piles by model testing. Cold Regions Science and Technology, 138: 98-107. DOI: 10.1016/j.coldregions.2017.03.008.

Wang TF, Liu JK, Tai BW,et al., 2017b. Frost jacking characteristics of screw piles in seasonally frozen regions based on thermo-mechanical simulations. Computers and Geotechnics, 91: 27-38. DOI:10.1016/j.compgeo.2017.06.018.

Wang X, 2005. Research of bored pile vertical carrying capacity in permafrost region. Harbin: Northeast Forestry University, pp.30-41.

Wang X, Jiang DJ, Liu DR,et al., 2013. Experimental study of bearing characteristics of large-diameter cast-in-place bored pile under non-refreezing condition in low-temperature permafrost ground.Chinese Journal of Rock Mechanics and Engineering, 32(9):1807-1812. DOI: 10.3969/j.issn.1000-6915.2013.09.011.

Wang XL, Chen PZ, Wu SH, 2003. Research and application of Qinghai-Tibet Railway pile foundation forms. Research and application of Qinghai-Tibet Railway pile foundation forms, (1): 66-68. DOI:10.3969/j.issn.1001-683X.2003.01.017.

Wen Z, Yu QH, Wang DY,et al., 2012. The risk evaluation of frost jacking for tower foundations along Qinghai-Tibetan transmission line and anti-heave measures. Proceedings of the Conference on Cold Regions Engineering. Quebec City, Canada. American Society of Civil Engineers, pp. 573-582.

Wen Z, Yu QH, Ma W,et al., 2013. Direct shear tests for mechanical characteristics of interface between Qinghai-Tibetan silt and fiberglass reinforced plastics. Rock and Soil Mechanics, 34(S2): 45-50.DOI: 10.16285/j.rsm.2013.s2.004..

Wen Z, Yu QH, Ma W,et al., 2016. Experimental investigation on the effect of fiberglass reinforced plastic cover on adfreeze bond strength. Cold Regions Science and Technology, 131: 108-115.DOI: 10.1016/j.coldregions.2016.07.009.

Wotherspoon LM, Sritharan S, Pender MJ,et al., 2010. Investigation on the impact of seasonally frozen soil on seismic response of bridge columns. Journal of Bridge Engineering, 15(5): 473-481.DOI: 10.1061/(ASCE)BE.1943-5592.0000075.

Wu YP, Guo CX, Pan WD,et al., 2004. Influences of refreezing process of ground on bearing capacity of single pile and bridge construction in permafrost. Chinese Journal of Rock Mechanics and Engineering, 23(24): 4229-4233. DOI: 10.3321/j.issn:1000-6915.2004.24.023.

Wu YP, Zhu YL, Guo CX,et al., 2005. Multifield coupling model and its applications for pile foundation in permafrost. Science in China Series D Earth Sciences, 48(7): 968-977.

Wu YP, Guo J, Guo CX,et al., 2010a. Exothermic process of cast-inplace pile foundation and its thermal agitation of the frozen ground under a long dry bridge on the Qinghai-Tibet Railway. Journal of Zhejiang University-Science A, 11(2): 88-96. DOI:10.1631/jzus.A0900522.

Wu YP, Liu Z, Wang N,et al., 2016. Experimental study of the thermal effect of pile bottom water on the rheological property of model pile-frozen soil. Chinese Journal of Rock Mechanics and Engineering, 35(S1): 3424-3431.

Wu ZJ, Che AL, Chen T,et al., 2010b. Test study and numerical analysis of seismic response of pile foundation of bridge at permafrost regions along Qinghai-Tibet Railroad. Rock and Soil Mechanics,31(11): 3516-3524. DOI: 10.3969/j.issn.1000-7598.2010.11.027.

Xiong F, Yang ZH, 2008. Effects of seasonally frozen soil on the seismic behavior of bridges. Cold Regions Science and Technology,54(1): 44-53. DOI: 10.1016/j.coldregions.2007.11.003.

Xiong W, Liu MG, Zhang QH,et al., 2009. Temperature distribution along piles in permafrost regions. Rock and Soil Mechanics, 30(6):1658-1664. DOI: 10.3969/j.issn.1000-7598.2009.06.023.

Xu CH, Xu XY, Qiu MG,et al., 2007. Numerical analysis of adfreezing force of engineering pile in permafrost. Journal of Harbin Institute of Technology, 39(4): 542-545. DOI: 10.3321/j.issn:0367-6234.2007.04.008.

Xu LJ, 2006. Analysis of single pile stress state in permafrost region.Harbin: Northeast Forestry University, pp. 22-39.

Yang Z, Zhang X, Niu FJ, 2014. Fluid-solid coupled finite element modeling of bridge pile foundations embedded in laterally spreading ground in cold regions. In: Proceedings of the Geo-Shanghai 2014. Shanghai, China: American Society of Civil Engineers, pp.689-699.

Yang ZH, Dutta U, Zhu DJ,et al., 2007. Seasonal frost effects on the soil-foundation-structure interaction system. Journal of Cold Regions Engineering, 21(4): 108-120. DOI: 10.1061/ASCE0887-381 X200721:4108.

Yang ZH, Li Q, Horazdovsky J,et al., 2017. Performance and design of laterally loaded piles in frozen ground. Journal of Geotechnical and Geoenvironmental Engineering, 143(5): 06016031. DOI:10.1061/(ASCE)GT.1943-5606.0001642.

Yu DZ, Cheng PF, Ji C,et al., 2015. Static and dynamic contrast test for bearing capacity of refrozen bridge pile foundation in patchy permafrost regions. Chinese Journal of Rock Mechanics and Engineering, 34(S1): 2845-2853. DOI: 10.13722/j.cnki.jrme.2014.1175.

Yu ZQ, Wang X, Jiang DJ, 2007. Experimental study on bearing features of bored pile under non-refreezing condition in permafrost region. In: Proceedings of International Conference on Transportation Engineering 2007. ASCE, pp. 3828-3834.

Yue CY, Ma ZY, Wang W,et al., 2015. Application of the interface elements between pile and soil. Journal of Engineering of Heilongjiang University, 6(3): 1-5. DOI: 10.13524/j.2095-008x.2015.03.037.

Zhang JM, Zhu YL, Zhang JY, 1999. Experimental study on settlement of model piles in frozen soil under dynamic loading. Science in China Series D: Earth Sciences, 42(S1): 30-37. DOI:10.1007/BF02878850.