The use of high-filled cut-and-cover tunnels (HFCCTs) provides an ideal solution for reclaiming usable land in northwestern China because of the unique landforms of the Loess Plateau. Different from traditional tunnel boring methods, the HFCCT is first constructed and then backfilled in layers in the trench. Due to the backfill above the cut-and-cover tunnel (CCT) is required in quantity, currently, the existed estimating methods of the earth pressure may not suit the high-filled constructions. The ability to estimate the load on such tunnels and high backfill projects is extremely important. Conceptually, the Marston-Spangler (M-S) theory for buried culverts may be used to estimate earth pressure on HFCCTs. However, the earth pressure would be quite different from that of buried culverts in terms of the backfill volume, cross-sectional shape of structure, and foundation conditions. This paper presents physical experiments and numerical investigations to verify the influence of cross-sectional shape (arch and rectangle) of CCTs, foundation settlement, and load reduction using expanded polystyrene (EPS) for the vertical earth pressure (VEP) distribution and vertical displacement around a CCT. The experimental results agree well with the numerical analysis results. Moreover, further comparisons were also made to analytical analysis based on M-S theory. The comparison results indicate that analytical solutions for buried culverts cannot be applied to HFCCTs directly. In order to obtain the earth pressure accurately, designers must consider many influential factors.

High-filled cut-and-cover tunnels (HFCCTs) can be used in northwestern China to reclaim usable land. Because of the ultra-high backfill used in HFCCTs, the stress on the cut-and-cover tunnel (CCT) will change. However, The calculation method adopted at present for earth pressure usually result in either overestimation or underestimation. Different from vertical earth pressure, lateral earth pressure (LEP) can help stabilize the structure of HFCCTs. The current methods for estimating LEP rely mainly on Rankine’s theory or empirical formulas. Using conventional methods has led to errors between actual values and estimated values, and such deviation increases with an increase in the backfill height over the CCT. Due to the complex soil-arching effect, the current methods used for calculating LEP for HFCCTs need to be modified. These modifications need to consider characteristics of the CCT, mechanical properties of the backfill soil, and the geometry of the landform. In this work, several influential factors were identified through numerical analysis using the finite element method and propose four corresponding coefficients of modification: k₀, the cross-sectional shape of the CCT; k₁, the mechanical properties of the backfill; k₂, the width of the CCT; and k₃, the coupled effect of the slope angle and ratio of the width of the valley floor to the width of the CCT. Because current methods and specifications may overestimate the LEP for HFCCTs in valleys, a general equation was formulated for modifying the LEP coefficient. Also, this general modification equation was verified that agrees well with the numerical analysis results for different cases.

The excessive load from the high embankments of high-filled cut-and-cover tunnels (HFCCTs) threatens the safety and stability of HFCCTs. Relatively low-compacted (RLC) soil can reduce such loads in the short term, but the long-term ability of RLC soil to reduce the load on HFCCTs is unknown. The changes in the soil arching effect that are resulted from soil creep needs to be considered carefully in HFCCT design. Three stages of analysis were conducted in this study: immediately after backfilling (0 month), during backfill creep, and after the soil deformation has stabilized (240 months/20 years). A finite difference program, FLAC3D, was employed to investigate the soil’s vertical earth pressure and displacement distribution around a cut-and-cover tunnel (CCT). The Burgers model was used to simulate the creep behavior of the soil. In addition, parametric studies for different dimensions and locations of RLC soil were conducted. The results show that the soil arching effect that is due to the inclusion of RLC soil works well initially to reduce the load but eventually disappears, which can result in a significant load rebound greater than 100 kPa and jeopardize the stability of the HFCCT.