Introduction
Constructing high-speed railways is an important strategy to speed up the process of railway modernization in China. To construct tracks that are smooth and highly precise, the settlement of the railway infrastructure must be stringently controlled [
1]. Meanwhile, the stability and deformation of the foundation should be considered when building a railway over soft soil, especially when constructing the embankment of the high-speed railway. After its construction, the embankment needs to be stabilized constantly, and the uneven settlement needs to be strictly confined. According to the regulation, the settlement after construction should be limited to less than 15 mm [
2].
The geosynthetic-reinforced and column-supported (GRCS) earth platform is a new type of composite foundation in which piles, geosynthetics and soil work together to support the upward loading [
3]. Compared to the embankment supported by piles without reinforcement, the GRCS earth platform with the combination of the pile and geosynthetics can effectively alleviate the uneven surface settlements [
4]. GRCS earth platforms can provide an economic and effective solution for embankments, retaining walls, storage tanks, etc. [
5] Therefore, the GRCS earth platform is used worldwide varying for highways, railways and architecture constructions [
6–
10].
Due to the complexity of the system, field test is considered as an effective way for investigating the GRCS earth platform [
11], and some field tests have been conducted to study the behavior of GRCS earth platforms. Chen et al. [
1] discussed three cases in which pile-supported embankments were used for constructing highways and proposed some available design standards and approaches for GRCS earth platforms which were verified by three cases. Liu et al. [
11] described a case history of a geogrid-reinforced and pile-supported highway embankment with a low area improvement ratio of 8.7%, the measured results are compared with computed result by carrying out three-dimensional fully coupled finite-element analysis. Le et al. [
12] found that the vertical loading of the embankment could be directly transferred to the piles by soil arching or indirectly transferred via the membrane effect of the geotextile. Lian et al. [
13] reported that transferring the load with a geogrid is better than transferring by soil arching. Chen et al. [
14] investigated three cases of GRCS earth platforms for highway construction in the eastern coastal region of China and the data indicated that significant soil arching in the embankment was observed. Furthermore, the earth pressures acting on the piles were much higher than the pressure acting on the soils between piles. Zheng et al. [
15] presented the technique of GRCS employed to treat the bridge approach in loess area, and found that the effect of load transfer is significant at the bridge approach and decreases gradually with the increase of distance from bridge abutment. Briançon and Simon [
16] describe a full-scale experiment of pile-supported embankments to provide a new design method, and the experiment shows the great influence of a granular platform reinforced with geosynthetics in load-transfer mechanisms.
Although some field tests have been performed, certain results on the bearing capacity, the pile-soil stress ratio, layered settlement and the work mechanism of geosynthetics do not agree [
13]. In addition, the current research on GRCS earth platforms is focused on normal embankments with a relatively small cross section resting on a specific commonplace soil foundation [
17–
20]. The GRCS earth platforms used to support a station yard of a high-speed railway resting on soft soil have never been studied, and there is no appropriate research or engineering principles can be used for the relevant application.
An approximately 15-month long field monitoring project was conducted based on the specific soft soil foundation in the district of the Chaoshan railway station. By taking the typical section (mileage: DK207+ 570) with relatively deep soft layers and a large cross section (the cross-sectional width of the subgrade is approximately 110 m) as the testing site, this paper analyzed the corresponding data in detail to investigate the bearing behavior and settlement mechanism of GRCS earth platforms resting on oversize-soft-soil. The conclusions will provide important references and a rational basis for the design and construction of GRCS earth platforms used to treat soft soil foundations; In addition, the results will further promote this new type of treatment method for soft foundations used in high-speed railway construction.
Brief introduction of the field test
Subsurface geotechnical condition
In the district of the Chaoshan Railway Station, the overlying strata, which are 10–32 m thick, are mainly composed of muck, muddy silty clay and silty sand deposited in the alluvial-lacustrine () and marine-continental () environment of the Quaternary. The underlying layer is composed of silty clay, silt, sand and fine rounded gravel formed in the alluvial environment of the Quaternary (), with a thickness of 30 – 60 m. The geological profile of the testing section and the CPT data are illustrated in Fig. 1.
Two types of soft soil layers can be found in the testing section, starting from the top downward: silt (III) and clay (V, VII and IX). The silt layer ranges from 7 to 24 m depth with a thickness of nearly 17 m, which is almost 28% of the entire thickness of the soil layers under investigation. Three layers of clay can be found in the testing section, and the total thickness of the three clay layers reaches approximately 30.8% of the thickness for the entire layer. The soft soil layer, which accounts for 58.8% of the entire depth investigated, contains many types of organic matter. In addition to the high proportion of organic matter, the rheology and thixotropy of the soft soil layer greatly affect the deformation and bearing behavior of the soil foundation.
The CPT data in Fig. 1 shows that the cone tip resistance of the silt layer between depths of 6 and 17 m ranges from 100 to 2100 kPa, and the corresponding sidewall skin friction varies from 5 to 30 kPa. However, the cone tip resistances of the soil layers from 0 to 6 m and from 17 to 35 m vary from 500 to 6000 kPa, and the relevant sidewall skin frictions range from 5 to 1000 kPa. These results imply that the compactness, the undrained shear strength, the compression modulus and the deformation modulus of the silt layer are less than those of the other layers between 0 and 35 m.
Test setup and instruments
The section in which the monitoring instruments were installed is located at the mileage of DK207+ 570 and the filling was conducted from July 31 to November 24, 2010, with some construction interruptions. The soft soil foundation of the testing section was treated using tube piles (type: PHC 500 A 100-12; diameter: 0.50 m; thickness: 0.1 m; concrete grade: C35), pile caps (concrete grade: C35; length and width: 1.6 m; height: 0.35 m), cushions, geogrids and ground beams (which adopted C35 concrete and were cast-in-place), as shown in Fig. 2. In the construction process of cushion layer, a layer of 0.2 m rubble was first paved, then a 0.05 m thickness of fine sand was filled; finally, the geogrid was installed above the fine sand layer. This construction process repeated for three times, and the final height of the cushion layer accounts for 0.7 m. The geogrids adopt the glass fiber with the type of GG8080 and the relevant design value of bidirectional tensile strength ranges from 80 to 100 kN /m. The fillings of the subgrade were constructed layer by layer to satisfy a specific compactness, and the filling types and corresponding parameters are listed in Table 1.
The earth pressure cell can be used to investigate the variation of pressure acting on the pile and soil as well as the pile-soil stress ratio in the loading process [
21]. In this work, it was adopted to measure the pressure acting on pile top and the soil between piles. A total of seventeen earth pressure cells were installed in the section; seven were installed on the pile cap and ten are installed in the soil between piles, as shown in Fig. 3.
To measure the axial force of piles, wire stress gauges were welded on the longitudinal bars along the pile body in different positions to measure the variation of the axial force during the construction and post-construction period; meanwhile, the skin friction of piles due to the consolidation of surrounding soil can be obtained as well. The tension of geogrid can be derived from the measured data of stretched geogrid recorded by displacement transducers; see Fig. 4 (a). The layered settlement was monitored using settlement gauges (Fig. 4 (b)), which were respectively arranged at the center of the embankment and the central lines of the railway tracks. The lateral displacement of the embankment and the pore water pressure were measured with inclinometers and pore pressure gauges, as shown in Figs. 4 (c) and (d), respectively.
Table 2 shows the details of the instruments used for testing. The specific locations of the embedded instruments are shown in Fig. 5.
A total of thirteen pore pressure gauges labeled PPG01 ‒ PPG13 were installed in the tested section, as shown in Fig. 5 (a). Among them, nine gauges were installed at a depth of 5 m, and 1 was installed at depths of 10, 15, 20 and 25 m to investigate the variation of the pore water pressure at different locations of the section.
As illustrated in Fig. 5 (b), six settlement tubes with 154 magnetic cores were installed. The magnetic core, which can measure the amount of settlement and uplift of the soil layers, was attached to the settlement tube every 2 – 3 m from the top downward.
In addition, four inclinometers were embedded at a depth of 30 m next to the slope toe of the embankment. These inclinometers can be used to measure the lateral displacement of the embankment, as shown in Fig. 5(b).
Results and analysis of the field testing
Earth pressure and pile-soil stress ratio
The data obtained from typical earth pressure cells are used to plot the curves of earth pressure versus filling height and time; see Fig. 6. The data of S3 (Fig. 3) represent the pressure acting on the soil in the center of four piles, and those of S10 indicate the pressure on the soil between two piles. The curves of the average pressure on the pile top and on the soil between piles are plotted against time on the bottom of Fig. 7, and the curves of pile-soil stress ratios are plotted against time on the bottom of Fig. 8. The curves for the filling height are plotted against time on the top of both figures.
Figures 6, 7 and 8 show the following trends:
1) The variation of the pressure acting on the pile top and on the soil between piles with filling height, which is similar to the variation of the pile-soil stress ratio with filling height, can be divided into two stages.
The first stage is the construction period, which lasts for 4 months (120 days). In this stage, the earth pressure on pile top grows linearly with increasing filling height. Beginning on the 37th day, the soil pressure between piles increases with the filling height at 5.6 kPa/d. From the 38th day to the 121st day, the soil pressure between piles decreases with increasing filling height at a speed of merely 0.1 kPa/d. In addition, the pile-soil stress ratio increases almost linearly with increasing filling height, and the corresponding rate of change is approximately 0.072/d.
The second stage is the stabilization phase in the post construction period, which lasts for 11 months (326 days). During this stage, the earth pressure on the pile top continues to increase, but the average rate of change decreases to approximately 0.083 kPa/d, which is much smaller than the rate of change during the first stage. The pressure acting on the soil between piles during the same period continues to decrease at a changing rate of 0.028 kPa/d. Additionally, the pile-soil stress ratio shows an increasing trend, and the corresponding rate of change is approximately 0.025/d which is smaller than the rate of change in the first stage.
2) The soil pressure between piles is much smaller than the pressure on the piles, and the pile-soil ratio is larger than 1 after the filling is completed. At the end of the monitoring work, the pile-soil ratio reaches 17, which indicates that the soil between the piles barely supports the load and that the main load is borne by the piles.
3) Around on the 35th day of filling, the maximum value of the soil pressure between the piles is reached, which indicates that a soil arch is formed at this specific filling height. At this point, the load borne by the soil between the piles begins to transfer to the piles. The specific filing height is approximately 1.3 m, i.e., the height of the soil arch is 1.3 m, which confirms the theory that the height of the soil arch is 1.1 to 1.5 times of the pile spacing [
22].
4) Fig. 4 demonstrates that the earth pressure measured by S3 in the center of four piles is 3.8 kPa, and the value measured by S10 between two piles is 21.2 kPa. Because the soil arch in the center of four piles is greater than which between two piles, the soil pressure in the center of the four piles is smaller than what in the center of two piles.
5) During the loading process from the 1st to the 121st day, the rate of change for the pile-soil stress ratio is relatively fast. Nevertheless, the rate of change becomes slow and then remains stable beginning on the 121st day, when the loading process is nearly complete, to the 449th day which is at the end of the monitoring work. Moreover, for a short period of time before and after the 300th day, the pile-soil stress ratio varies from 12 to 17 due to the continuous adjustment of the cushions on the load borne by the piles and the soil between the piles.
Behavior of geogrid
A total of five flexible displacement transducers with the same original length of 22 mm were installed in the testing section, and they were labeled as FDT1, FDT2, FDT3, FDT4 and FDT5, respectively. All of them were banded below the first layer of the geogrid. The data obtained by FDT2 and FDT3 (which were located on the pile top and in the soil between piles, respectively) were selected to plot the elongation of geogrid with the filling height as a function of time, as illustrated in Fig. 9.
The following trends can be observed:
1) In general, the amount that the geogrid stretched increases with the filling height, especially for FDT3, which is located in the soil between piles. The elongation of the geogrid in the soil between piles is larger than that on the pile top. When the monitoring work ended, i.e., on the 320th day after the filling began, the elongation measured by FDT2 on the pile top is 1.11 mm; however, the elongation measured by FDT3 in the soil between piles is 3.5 mm. The former is approximately three times larger than the latter.
2) The elongation of the geogrid in the soil between piles increases slowly in the early stage of filling process due to the tensile strength and creep behavior of the geogrid. Subsequently, the rate of change for elongation increases gradually then grows relatively fast, and eventually returns to a moderate level.
The average changing rate of elongation between the 1st and the 31st day of filling is 0.012 mm/d, and the average changing rate increases to 0.03 mm/d between the 31st and the 101st day. The rate of change for elongation in the later days of the filling period is clearly larger than that in the earlier days. As the filling develops, the soil arching is gradually generated; subsequently, the loading borne by the geogrid becomes relatively small, resulting in a smaller rate of change for elongation.
3) From the 250th to the 449th day of monitoring project, the elongation of the geogrid in the soil between piles fluctuated slightly. Because the differential settlement between the pile and the soil occurs after the load is applied, the load borne by the soil between piles is gradually diverted to the piles through the cushion. The fluctuation of elongation is caused by the cushion for its constant adjustment on the load supported by the soil and the pile.
4) The elongation rate of the geogrid can be calculated from the values of corresponding elongation, and the tension of the geogrid can subsequently be obtained, the results are listed in Table 3.
It can be inferred that the elongation rate and the tension of the geogrid located in the soil between piles are both larger than those on the pile top, but the overall elongation and tension of the geogrid are not necessarily large.
Axial force and skin friction of piles
Axial force
By arranging the wire stress gauges along the testing piles, the distributions of the axial force and skin friction can be measured to determine the load transfer process. Therefore, two testing piles, numbered TP1 and TP2, from the left to the right of the embankment are arranged, and 13 wire stress gauges are installed along each pile; see Fig. 5(a).
The changes in axial force are measured by TP1 and TP2, and the data versus depth are plotted in Figs. 10 and 11, respectively. The following can be observed in these figures:
1) Under different filling heights, the distribution of the axial force in the tube piles appears to be nonlinear, i.e., the axial force increases nonlinearly with depth, and the distribution of axial force is uneven (the axial force is small on the bottom and the top part, but large in the middle part).
2) The variation of the axial force on the pile body can be divided into three segments from the top downward: the upper, the middle, and the lower segment.
The upper segment is located at depths of 3 – 6 m along the pile body, and the axial force increases continuously in this segment from the top downward. In addition, the slope of the axial force curve is relatively large. This indicates that the load transfer of the axial force is relatively fast, mainly because the soil surrounding the piles in this segment is coarse sand with high strength, making the skin friction of the piles relatively large. Therefore, the load transfer proceeds smoothly.
The middle segment is from depth of 6 ‒ 18 m along the pile body, and the axial force continues to increase in this segment from top to bottom. The slope of the axial force curve becomes smaller than the upper segment, and the load transfer becomes relatively slow because the soil layer surrounding the piles in this segment is a silt layer with low strength.
In the lower segment, from depths of 18 to 35 m along the pile body, the axial force shows a decreasing trend from the top downward. The slope of the axial force curve from 18 to 24 m is small, and the load transfer is slow accordingly. However, the slope of the axial force curve is large, and the load transfer becomes fast from 24 to 35 m. The average load transfer speed at depths from 24 to 35 m of the pile body is smaller than that of the upper segment, but it is much larger than that of the middle segment. This is mainly because the soil layers surrounding the piles in this specific segment are formed by fine sand, silt and coarse sand with relatively high strength, making the skin friction of the piles larger than that of the middle segment in the silt layer.
3) The rate of change for the axial force with time at the bottom and on the top of piles is fairly small. However, the corresponding rate of change is relatively large in the middle part of piles. The maximum average changing rate of the axial force with time measured by TP1 and TP2 both appear at a depth of 18 m along the pile body. The relevant values measured by TP1 and TP2 are 0.92 and 0.89 kN/d, which are 23 and 22.35 times larger than the value measured on the top of pile, respectively. The increased loading has a greater effect on the middle part of the tube piles and less of an effect on the pile heads.
4) According to the load transfer properties reflected by the axial forces and the skin friction of the testing piles, the pile segments located in coarse sand, medium sand and silt bear relatively large loads. In addition, the resistance at the bottom of piles is small but not zero, which indicates that the load bearing behavior of the tube piles in this section, appears to be an end-bearing friction pile.
Skin friction
The skin friction is derived from the measured axial force of piles, and the average skin friction of each pile segment is calculated from the measured axial force divided by the area of the side surface. The distribution of the skin friction of each pile varies with depth as illustrated in Figs. 12 and 13.
The following observation can be drawn:
1) According to the distribution of the skin friction shown in the figures above, the pile body can be divided into three segments. In the first segment, the skin friction is relatively large and appears to be negative for the two piles at depths of 3 to 6 m. The skin friction of the second segment is relatively small and negative, ranging from 6 to 18 m in depth. In the third segment, the skin friction is positive and the depths vary greatly from 18 to 35 m. In addition, because the effective depths measured from TP1 and TP2 are both 35 m, the effective depth of the skin friction is 35 m for those two piles.
2) The value of the skin friction is related to the properties of the soil layers. The first segments of the two piles are located in the coarse sand layer, and the skin friction is negative and relatively large. The skin frictions measured by TP1 and TP2 are 18.78 and 25.97 kPa in the first segment on January 5, 2010, respectively. In the second segment, the two piles are located in silt layer and the corresponding skin friction is negative and relatively small. In this segment, the skin frictions of TP1 and TP2 range from 1.7 to 2.17 kPa and 1.66 to 5.84 kPa on January 5, 2010, respectively. The third segment is located in layers of silt, coarse sand, fine sand and clay, and the relevant skin friction is positive and relatively large. The skin frictions of TP1 and TP2 in the third segment range from 1.21 to 15.89 kPa and 1.91 to 20.21 kPa, respectively.
3) With increasing time and load, the skin friction on each segment of the piles generally shows a gradually increasing trend. For example, the skin friction of the piles at depths of 31 to 35 m, as measured by TP2, is 10.31 kPa on June 15, 2009, and increases to 20.21 kPa on January 15, 2010.
4) The dividing point for positive and negative skin friction or the point of zero skin friction value is considered as the neutral point. The diagrams show that the neutral point occurs in both of the testing piles and is located at the depth of 18 m along the pile body.
Pore water pressure
The work of monitoring the pore water pressure began on July 16, 2009 and ended on November 24, 2010, for 16 months in total. To facilitate the comparison of the data measured by pore pressure gauges, the data measured by PPG02, PPG04- PPG07 (see Fig. 5 (a)) located at the same depth (PPG01, PPG03, PPG08 and PPG09 were damaged during construction) are plotted in Fig. 14, and the data from PPG10- PPG13 embedded at different depths are plotted in Fig. 15.
The following can be inferred from these two diagrams:
1) The pore water pressure increases immediately when filling began on August 31, 2009; therefore, the excess pore water pressure is generated at the very beginning. Gauges PPG05 and PPG13 measured 0.0015 and 0.0014 MPa, respectively, at the start of the filling process.
2) From the beginning of filling to the construction period after filling, the pore water pressure increases during the filling stage and decreases during the interruptions in construction. For example, during the period from the 111th to the 117th day, the pore water pressure measured by PPG04 increases from 0.03967 to 0.04017 MPa as the filling height increases from 4.168 to 4.662 m. However, the pore water pressure decreases from 0.04017 to 0.03969 MPa during the intermission from the 117th to the 120th day.
3) When the filling process is completed, the filling height is 5.337 m. The pore water pressure measured by PPG05 located a depth of 5 m is 0.01368 MPa while the pore water pressure measured by PPG12 (located at a depth of 20 m) and by PPG13 (located at 25 m) are 0.01208 and 0.01006 MPa, respectively. The pore water pressure shows a decreasing trend with increasing depth.
4) On the 143rd day, the filling work was finished. The dissipation process of excess pore water pressure, which is monitored by PPG02 ‒ PPG09 at a depth of 5 m in the coarse sand layer, took approximately 20 days. However, the process to dissipate the excess pore water pressure, as monitored by PPG10 at a depth and PPG11 at a depth of 15 m in the silt layer, took nearly 100 days. These data indicate that the release speed of the excess water pressure has a close relationship with the physical and mechanical properties of the soil layers. Specifically, the dissipation of excess pore water pressure proceeds quickly in the coarse sand layer and the medium sand layers, and it proceeds slowly in the silt layer and the silty soil layers.
Layered settlement
Variation in the settlement with filling height and time
Based on the data measured by the magnetic cores installed in different soil layers, the curves of layered settlement with filling height as a function of time are plotted in Figs. 16, 17, 18, 19, 20 and 21.
Figures 16, 17, 18, 19, 20 and 21 show the following trends:
1) The settlement of each soil layer at different depth increases significantly with increased filling height and time. When the filling height is relatively small, the settlement of each soil layer at different depths is also small. However, the corresponding settlement increases as the filling height develops.
2) When the filling speed is relatively fast, The rate of change for the settlement of each soil layer at different depths begins to increase, and the slope of the settlement curves increases abruptly, i.e., a steep increase occurs in the layered settlement. As an example, the data measured by ST1 at 1.625 m show that the filling height developed from 0.6 to 1.384 m between the 10th and the 25th day. During this period, the settlement of the soil layer at a depth of 1.625 m increased from 36 to 90 mm at a relatively high rate of change with a significant decrease in the settlement curves, as shown in Fig. 16.
3) During the filling interval, the settlement of each soil layer at different depth fluctuates or increases slightly, and the increase rate in these intervals is significantly smaller than that when the filling is carried on.
4) The layered settlement curves show that the settlement of each soil layer increases abruptly when a large load is applied in a short time at the beginning; the settlement then decreases at different rates over the next 10 days. In addition, these settlement curves fluctuant during the entire filling process, and this phenomenon is observed in all the testing tube piles. As an example, the monitored settlement data of ST1, at a depth of 1.625 m, show that the settlement increases from 36 to 90 mm between the 10th and the 25th day; the settlement then decreases to 82 mm from the 25th to the 30th day.
5) The variation of the settlement obtained by the magnetic cores at different depths in the same settlement tube is basically the same, and the settlement is inversely proportional to the embedded depth of the magnetic cores, i.e., the settlement of the soil layer at a shallow location is larger than that at a deep location.
6) The settlement of the soft soil layers, such as a silt layer, is larger than that of other layers. For instance, the settlements of the silt layer at depths of 9.471 and 13.457 m measured by ST1 are approximately 178 and 158 mm, respectively.
7) After the filling work is finished, the weight of the embankment remains essentially constant, and the settlement of each soil layer changes steady. During entire monitoring process, the rate of change for the settlement is almost directly proportional to the filling speed.
Analysis of the varying settlement
The entire settlement monitoring work lasts 429 days in all. The data from the 45th day (October 23, 2009; during the filling period), the 113th day (December 31, 2009; the day when the filling is finished) and the 142nd day (January 1, 2010; time after the filling is completed) were selected to plot settlement contours, as shown in Figs. 22, 23 and 24.
According to the contours of the foundation settlement on these three days, the following can be inferred:
1) The settlement of the same soil layer shows an increasing trend with time. For instance, the settlement, 10 m left away from the subgrade center, is 110 mm on the 45th day and increases to 180 mm on the 113th day. The settlement of the subgrade center is 160 mm on the 113th day and changes to 180 mm on the 142nd day. The settlement difference between the 113th and the 142nd day is relatively small because the load remains constant after the 113th day, when the filling is finished, and the settlement appears stable with time.
2) Differential settlement occurs at different locations along the cross-sectional direction of the subgrade. On the 142nd day, the settlement of the subgrade center at a depth of 10 m is 180 mm, while the settlements 30 m away from the subgrade center on the left and right side at depths of 10 m are 170 and 160 mm, respectively. The relatively large cross-sectional width (approximately 110 m) and the different distributions of the soil layers at different locations are likely the main reasons for the generation of differential settlement.
3) During the construction period and the post construction period, the differential settlement changes depending on time and locations. On the 45th day, the maximum settlement of the embankment occurs in the center of the subgrade (see Fig. 22), and the settlement shows a significant decrease from the subgrade center to the edge. On the 113th day, the maximum settlement appears to be 20 m away from the subgrade center on the left side, as shown in Fig. 23. After almost one month, the location of the maximum settlement eventually remains at place which is 16–18 m away from the subgrade center on the left side, which is similar to the result on the 113th day.
Lateral displacement
A total of 4 inclinometer tubes are symmetrically installed at the locations of slope toe and place which is 4.5 m away from the slope toe. The inclinometer tubes are labeled IT1, IT2, IT3 and IT4 from the left to the right, as shown in Fig. 5 (b). To simplify the analysis, only the data measured by IT3 and IT4 were investigated, and the corresponding curves for the lateral displacement varying with time are plotted in Figs. 25 and 26. In addition, the curves for the maximum lateral displacement varying with load and time are plotted in Fig. 27.
Figures 25, 26 and 27 show following trends:
1) The lateral displacement of the embankment begins in the early stage of the filling period and shows a decreasing trend with increased depth, i.e., the lateral displacement of the embankment at shallow depths is relatively large. For instance, on the 449th day (October 9, 2010), when the monitoring work was finished, the lateral displacement measured by IT4 was approximately 20 mm at depth ranging from 0 to 5 m, and fluctuated around 0 mm at depths of 15–17.5 m (see Fig. 26).
2) As the filling height and the consolidation of soil proceeding, the lateral displacement of the embankment increases accordingly. The changing rate of the lateral displacement during the construction period is larger than that during the post construction period. For example, the average changing rate of the lateral displacement observed by IT4 is approximately 0.12 mm/d during the construction period, but it decreases to 0.09 mm/d in the post construction period, as shown in Fig. 26.
3) The lateral displacement of the embankments under loading varies in different soil layers. Generally, the lateral displacement of the embankment in soft soil layers, such as the silt layer or the silty clay layer (Fig. 1), is relatively large. Furthermore, the lateral displacement around the dividing line of two soil layers from the ground surface to a depth of 10 m is particularly large. Specifically, the maximum lateral displacement measured by IT3 occurs at a depth of 6 m, which is next to the boundary between the coarse sand layer and the silt layer (Fig. 25). The maximum lateral displacement measured by IT4 occurs at a depth of 2.5 m which is between the ground surface and the silt layer (Fig. 26).
4) From the 229th day to the 244th day, the subgrade is broadened considerably, i.e., some additional loading is applied on the embankment in a short time. This leads to the maximum lateral settlement measured by IT3 increased another 12 mm abruptly as shown in Fig. 27.
Conclusion
Based on the field testing of GRCS earth platforms for the Chaoshan high-speed railway station, the bearing behavior and the settlement mechanism of GRCS earth platforms resting on soft soil were investigated. The following conclusions can be drawn:
1) In the early stage of filling work, the pressure acting on the soil between piles and on the pile top increase drastically, however the changing rate for the pressure on the pile top is larger than that on the soil between piles. When the filling reaches a certain value, the development of the earth pressures both on the piles and the soils shows that there was a significant soil arching in the embankment, and the earth pressure in the center of four piles is larger than that between two piles. The pile-soil stress ratio increases with time and loading, and it fluctuates due to the constant adjustment of the cushion. When the filling is finished, the pile-soil stress ratio reaches 17, which indicates that the majority of the load is borne by the tube pile.
2) The elongation of the geogrid increases with the filling height, and the elongation in the soil between piles is larger than that on the pile top. The elongation rate and tension of the geogrid located in the soil between piles are both larger than the corresponding values on the pile top, and the overall elongation length and tension of the geogrid are not necessarily large.
3) Both the axial force and the skin friction of piles increase with time and load. In addition, the force and the friction are closely related to the properties of the soil layers. In the silt layer and the silty soil layer with low strength, the skin friction is relatively small, and the axial force can be quickly transferred. In the fine sand layer, the medium sand layer and the clay layer with relatively high strength, however the skin friction is relatively large, and the axial force is transferred slowly. In addition, the tube piles used to strengthen the soft soil in this section appear to be the end-bearing friction piles.
4) The pore water pressure increases slightly with the filling height because the load borne by the soil between the piles is relatively small. The excess pore water pressure is released quickly in the coarse sand layer and the medium sand layers, and the pressure is released slowly in the silt layer and the silty soil layer. The increased load greatly affects the pore water pressure in the shallow location but has a relatively small influence in the deep location.
5) The changing rate of the layered settlement is directly proportional to the filling speed. The settlement increases abruptly when filling is conducted in a short time; the settlement then develops gradually and eventually decreases after a certain time. The properties of the soil layers also affect the settlement value; for instance, the settlement in soft soil layers is relatively large. Differential settlement occurs at different locations along the cross-sectional direction of the subgrade. Furthermore, the difference in settlement depends on the filling period (before or after the time when the filling is completed) and the measuring location.
6) As the fill height and the consolidation of soil proceeding, the lateral displacement of the embankment increases accordingly. The changing rate of the lateral displacement during the construction period is larger than that during the post construction period. The lateral displacement of the embankments under loading varies in different soil layers, and the lateral displacement of the embankment in soft soil layers is relatively large.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(2934KB)
