Y.-L. Wei’s scientific contributions

The experimental section is located in the class IV dead-weight collapsible loess site and is composed of three different kinds of loess foundation treatment measures. The transition sections without foundation treatment exist between them. All 22 m depth of collapsible loess is treated in zone of piles with down hole dynamic compaction. The upper 15 m depth of collapsible loess is treated in zone of cement soil compaction piles. The upper 6 m depth of collapsible loess is treated in zone of dynamic compaction. The results show that the collapsibility of loess within treatment depth has been completely eliminated. Their bearing capacity is higher than the standard value. The composite foundation settlements of piles with down hole dynamic compaction and cement soil compaction piles are less than 15 mm after construction for high-speed railway; but the dynamic compaction foundation can not meet the same requirement. In zone of pile with down hole dynamic compaction, the minimum and average coefficients of compaction are respectively more than the standard values of 0.88 and 0.93; but the average coefficient of compaction and compression modulus of piles are respectively less than the standard values of 0.97 and 100 MPa. In zone of piles with cement-soil compaction piles, the minimum and average coefficients of compaction, or the average coefficient of compaction and compression modulus of piles are also less than the same standard values. Average compression modulus of dynamic foundation is less than the standard value of 15 MPa. The reasonable values of test standard need further study.

Eight typical sites for water immersion tests of collapsibility of loess foundation were determined by combination of the distribution characteristics of loess along the Zhengzhou-Xi'an Passenger Dedicated Line. The test results show as follows: The two water immersion test sites in the section between Zhengzhou and Mianchi exhibit the maximum self-weight collapse settlement of 0.47 cm to 2.65 cm, belonging to class II of non self-weight collapsible sites; the six water immersion test sites in the section between Mianchi and Xi'an exhibit the maximum self-weight collapse settlement of 10.4cm to 160.3 cm, belonging to class II to class IV of self-weight collapsible sites; self-weight collapsibility of loess basically obeys law of gradual strengthening from east to west; for the low-limit depth of self-weight collapsible loess, the values measured at sites are generally less than the calculated values of samples; there is a relationship of the former controlling the latter in sequence with regard to the immersion area, scope of self-weight collapse deformation and range of cracks occurrence. The loess collapsibility of the Zhengzhou-Xi'an Passenger Dedicated Line was evaluated by sections and the foundation treatment depth, the design depth of negative friction of pile foundation and the widths of roadbed waterproof and drainage are recommeded.

The impact rolling is a kind of new technology in the area of ground treatment. Compared with the traditional rolling machines, it is more effective, the cost is lower, and it can treat thicker soil, especially it has obvious advantages in soft groud treatment. A new railway is under construction, most of which are embankment on soft ground soil. The bearing capacity of the shallow natural foundation is less than 150 kPa, and it should be treated. Based on the pilot projects, the process and parameters of ground treatment are studied, and effects of ground treatment are tested. The conclusions may provide technical supports for the application of impact rolling in treatment of the whole line.

The research results based on Qinghai-Tibet railway project and relevant test data indicate that the main factors, which influence the artificial upper table for embankment of Qinghai-Tibet railway in permafrost region, are in turn the temperature, the mean annual ground temperature, the fill type, the embankment height, the slope orientation and wind, and the construction season, etc. The temperature affects the artificial permafrost table through the downward dissemination of embankment surface temperature. The mean annual ground temperature as a cold source underneath is also playing a certain role in it from bottom to top. The coarse-grained soil as embankment fill can meet the basic condition of heat transfer theory to uplift artificial permafrost table. It is a positive correlation between the embankment height and the uplift value of artificial permafrost table in a reasonable range of height. The embankment, which runs from east to west, easily forms an increasing asymmetrical artificial permafrost table, but which runs from north to south, easily forms an increasing symmetrical one. The high temperature interlayer has been formed in basement strata of the embankment constructed during the warm season. It affects the restoration of embankment ground temperature field.

The Zhengzhou-Xi'an passenger dedicated line allows 15 mm settlement of embankment after acceptance. The settlement control of ground treatment in collapsible loess is a key technical problem. Therefore, the settlement of an embankment construction on ground treatment with different measures is observed and researched experimentally in a selected area of typical collapsible loess. The results show that the compression settlement of embankment body is only 3 mm and have already stabilized in 15 days after completion of embankment. The embankment settlement is primarily controlled by the foundation settlement which is determined by the settlement of soils below three different kinds of ground treatment layer. A section of pile thrusted-expanded in column-hammer which treats all 22 m depth of collapsible loess is best, a section of cement soil pile which treats 15 m depth of collapsible loess is better, a section of dynamic compaction which treats 6 m depth of collapsible loess is worst in the settlement, ground treatment and immersion effect. It is very essential that the reasonable waterproof and drainage measures are taken to prevent water from immersing the vicinity of slope foot. The average value of results which are predicted by the hyperbolic method, three-point method and Asaoka method is used as the final settlement of embankment. The constant preloading can be removed to lay ballastless track after 5 months in the section of pile thrusted-expanded in column-hammer and after 8 months in the section of cement soil pile. But it may need to continue for a long time in the section of dynamic compaction.