^{1,2}

^{2}

^{2}

^{1}

^{1}

^{2}

Cement fly ash gravel (CFG) pile composite foundation is an effective and economic foundation treatment approach, which is significant to building foundation, subgrade construction, and so forth. The present paper aims at investigating the settlement behaviors of saturated tailings dam soft ground under CFG pile composite foundation treatment, in which FEM and laboratory model test were utilized. The proposed findings demonstrate that CFG pile treatment is effective in reinforcing saturated tailings dam and loading has little influence on settlement of soil between piles. The variation of soil between piles settlement in FEM has a good agreement with the laboratory model test. Additionally, the cushion deformation modulus has a small effect on the composite foundation settlement, although the cushion thickness will generate certain influence on the settlement distribution of the composite foundation.

With the rapid development of highway construction in China, soft ground treatment, as a limiting factor for construction schedule, costs, and engineering quality, has received more and more attentions. CFG pile composite foundation was extensively used in ground treatment due to its large extent in bearing capacity improvement, being applicable to various ground cases, rapid construction, low engineering costs, and so forth [

To date, an increasing number of analytical, experimental, and numerical approaches have been developed to investigate the behaviors of CFG pile composite foundation under flexible foundation. In terms of theoretical analysis, for example, He et al. [

Wangzhuangbao-Fanshi expressway is located in Shanxi, China; the route has nine sections drilling through or passing near saturated tailing dam area. Because the bearing capacity could not meet design requirements in the nine sections, three treatments (cushion, cement soil plus cushion, and CFG pile composite foundation) were employed in different sections, and the CFG pile composite foundation was applied at the sites between K55 + 650 and K55 + 770 sections of Yuehong Magnetic Plant saturated tailings dam (the bottom of tailing dam is “V” shaped gully, and the bottom layer is loess layer), which showed the best reinforcement effect among the three treatments. The expressway passed through the middle of tailings reservoir in the form of filling subgrade, in which CFG pile composite foundation was used to reinforce saturated tailings sand foundation (Figure

Cross section and

Cross section diagram (unit: cm)

The dimensions of the FE model are of 14 m × 10 m × 31.5 m. Displacements are restricted at the model boundaries in the normal direction to their respective planes. From the bottom to the top of the FE model, the soil profiles consist of loess layer, saturated tailings sand layer, gravel cushion, and subgrade filling layer, respectively, in which the depths of saturated tailings sand and loess layers are 25 m in total, gravel cushion is 3 m, and subgrade filling is 3.5 m. The interface of all soil layers was simplified to be plane, and soil was assumed to be solid element which obeys elastic perfectly plastic Mohr-Coulomb yield criterion [^{2}, 100 kN/m^{2}, and 150 kN/m^{2}, respectively, based on the

Physical parameters of soils and CFG piles [

Number | Material types | Modulus/MPa | Poisson’s ratio | Soil unit weight/(kN/m^{3}) |
Cohesion/kPa | Angle of internal friction/(°) |
---|---|---|---|---|---|---|

1 | CFG pile | 1600 | 0.25 | 21.5 | 900 | 35 |

2 | Saturated tailings sand | 2 | 0.32 | 18.7 | 7 | 40 |

3 | Loess | 40 | 0.25 | 20 | 35 | 45 |

4 | Gravel cushion | 140 | 0.16 | 20 | 0 | 36 |

5 | Filling layer | 100 | 0.2 | 19 | 15 | 20 |

FE model of composite foundation and pile-soil contact element.

Integral model

Pile-soil contact element

Distribution of soil settlement in different soil layers.

Case of subgrade filling construction procedure

Case of loading of 150 kPa

Variation of soils settlement with depths.

Variation of soil settlement in horizontal direction.

Variation of soil settlement with subgrade transverse distance.

Variation of soils settlement under different loads.

Variation of soil settlement with time.

Xing [

Laboratory model test of CFG pile (unit: m).

The CFG model pile was made of stone chip, sand, gravel, fly ash, cement, and water. The length of the CFG pile is 1.5 m, diameter is 0.05 m, and 35 CFG piles in total were employed in the model test. Layout plan of CFG pile in the model test is shown in Figure

Layout diagram of CFG piles (unit: m).

Laboratory model test model [

Model trough

CFG model pile

The results of the FEM and laboratory model test on the CFG pile composite foundation settlement demonstrate that the settlement of various soil layers is consistent and stable at different pile positions. Therefore, the soil between piles at central pile (with maximum settlement) was selected to carry out the comparative analysis on the results between the FEM and laboratory model test which investigated the variation of composite foundation settlement with load, depth, and time. And settlement value of the laboratory model test results was multiplied by settlement similarity ratio

Settlement variation of soil between piles with load is illustrated in Figure

Comparison of interpile soil settlements under different loads between FEM and model test.

Figure

Comparison of interpile soil settlements at different depths between FEM and model test.

Presented in Figure

Comparison of interpile soil settlements with time between FEM and model test.

The variation of settlement of soil between piles in FEM exerts a good agreement with the laboratory model test; thus, to a certain extent, the FEM results can better reflect the settlement changes in practical engineering. The results from FEM analysis show that the cushion has the maximum settlement, and the average value of settlement difference between cushion and soil between piles reaches about 7 cm. Hence, it is quite important to improve cushion performance to reduce the settlement of composite foundation. And there have been some studies which discussed and introduced the effects of the cushion on the properties of composite foundation [

Pile-cushion system is important for the success of CFG pile composite foundation [

The deformation modulus of cushion in the aforementioned model is

Settlement distribution of composite foundation under different deformation modulus.

Distribution of composite foundation settlement under different deformation modulus with depth.

Variation of the settlement at central pile position under different modulus.

This section aims to explore the influence of the cushion thickness on composite foundation settlement. Deformation modulus of cushion

Settlement distribution of composite foundation with different thickness of cushion.

Distribution of composite foundation settlement under different thickness of cushion with depth.

In summary, the settlement distribution of composite foundation in different soil layers is very homogeneous when the thickness of cushion is around 1 m, while thick cushion will limit the development of the CFG pile bearing capacity. Therefore, cushion thickness of about 1 m is reasonable when CFG pile composite foundation is used in the treatment of saturated tailings dam.

The current paper presented FE model to study settlement behaviors of CFG pile composite foundation. In this model, the variation of the various soil layers settlement was investigated and the results were validated through comparison with the laboratory model test results. In addition, the effects of cushion on the CFG composite foundation settlement were studied. According to the results of settlement studies conducted in this research, some of the main findings are summarized as follows:

According to the FEM results, it is confirmed that the distribution of various soil layers settlement is homogeneous and stable except for the interface of soil layers. After loading, settlement rate of soil between piles (saturated tailings sand) is slower than other soil layers and tends to be stable more rapidly. The increasing range of the settlement in saturated tailings sand is relatively smaller than other soil layers, which is around 2.1 cm.

The variation of settlement of soil between piles in FEM is consistent with laboratory model test with depth, load, and time. Furthermore, in both cases, the results demonstrate that the various soil layers settlement of CFG pile composite foundation behavior is consistent and stable, which signifies that saturated tailing dam has obtained a good reinforcement effect after treatment of CFG pile composite foundation.

When the cushion deformation modulus is around 0–100 MPa, increasing the deformation modulus will exert a certain influence on the subgrade filling and cushion. However, there is small effect on the composite foundation settlement when it is more than 100 MPa. The thickness of cushion has a relatively large influence on the settlement distribution of the composite foundation. When the thickness of cushion is around 1 m, the settlement has approximately linear decrease with depth. At the same time, the settlement distribution of composite foundation in various layers is more homogeneous than the case of the other cushion thickness. Therefore, it is suggested that modulus of deformation of about 100 MPa and cushion thickness of about 1 m are reasonable, respectively.

Based on the results of FEM, cushion has the maximum settlement at different loading cases which is 7 cm larger than that of soil between piles on average. In addition, the settlement control effects are not quite obvious by means of changing the modulus and thickness of the cushion. Therefore, in order to further optimize the design scheme for CFG pile composite foundation and improve the performance characteristics of composite foundation, further experimental studies should be conducted on the new cushion material and the novel connecting type among cushion, pile, and soil of CFG pile composite foundation under flexible foundation.

The authors declare that there is no conflict of interests regarding the publication of this paper.

This work is financially supported by the Special Fund for Basic Scientific Research of Central Colleges of Chang’an University (Grant no. 31082116011), the Key Industrial Research Project of Shaanxi Provincial Science and Technology Department (Grant no. 2015GY185), and the Integrated Innovation Project of Shaanxi Provincial Science and Technology Department (Grant no. 2015KTZDGY01-05-02).