Passive earth pressure of narrow backfill considering seismic-unsaturated seepage multi-field coupling effect

Abstract

Narrow backfill earth pressure estimation is applied to study the stability of supporting structures in the vicinity of existing buildings. Previous narrow backfill earth pressure studies have neglected seismic-unsaturated seepage multi-field coupling, resulting in inaccurate estimates. To address these deficiencies, this paper proposed a calculation method for seismic passive earth pressure in unsaturated narrow backfill, based on inclined thin-layer units. It considers the interlayer shear stress, arching effect, and the multi-field coupling of seismic-unsaturated seepage. Additionally, this paper includes a parametric sensitivity analysis. The outcomes indicate that the earthquake passive ground pressure of unsaturated narrow backfill can be reduced by increasing the aspect ratio, seismic acceleration coefficient, and unsaturation parameter α. It can also be reduced by decreasing the effective interior friction angle, soil cohesion, wall-earth friction angle, and vertical discharge. Furthermore, for any width soil, lowering the elevation of the action point of passive thrust can be attained by raising the effective interior friction angle, wall-earth friction angle, and unsaturation parameter α. Reducing soil cohesion, seismic acceleration coefficient, and vertical discharge can also lower the height of the application point of passive thrust.

1. Introduction

Unsaturated seepage caused by rainfall will increase the matrix suction of soil mass, leading to slope instability [1, 2]. Presently, there are two major theories to discuss the implication of matric suction on the shearing strength of unsaturated soil. One is the Mohr-Coulomb modification form proposed by FREDLUND et al [3] based on the two stress state variables of effective normal stress and matrix suction. The Mohr-Coulomb strength criterion for three-dimensional damage surfaces can be plotted based on the above two stress variables. Another model is the generalized form of the efficient pressure equations proposed by BISHOP et al [4, 5]. The distinction between the two lies in their treatment of matrix suction when assessing shear strength. The former method incorporates the efficient stress parameter χ to quantify the influence of suction stress, while the latter method incorporates the suction angle φ^b to account for the effect of substrate suction. However, the use of the efficient pressure factor χ only corrects for the magnitude of substrate suction without clear physical meaning [6]. To address the aforementioned theoretical shortcomings, LU et al [7] directly introduced suction stress into effective stress and proposed an efficient pressure approach built on stress-absorption. This approach does not need to draw shearing strength standards and additional factors, thereby simplifying the calculations. In addition, LU et al [8] also introduced apparent cohesion to represent the variation of absorption stress. The aforementioned studies on unsaturated earth pressure described the effect of matrix suction systematically, but most of them ignored the effect of earthquakes [9−11].

Seismic hazards seriously threaten the safety of structures and cause destabilizing damage [12]. Ensuring the normal functionality of bridges, retaining walls, and other structures under seismic conditions is a complex challenge that needs to be addressed. Until now, the most popular and simple analysis framework for assessing the earthquake passive soil pressure has seemed to be the Mononobe-Okabe method [13]. In the past few years, researchers have proposed limit equilibrium solutions for soil resistance under seismic conditions [14, 15]. The above static assumptions are easier to implement, but they do not consider arching effects and dynamic time history properties and have limited applicability to rough wall backs [16]. Therefore, the pseudo-dynamic solution has been proposed to solve the simplified seismic force problem, taking into account the characteristics of the time history, vibration period, and amplification effects [17, 18]. This method assumes that the horizontal and vertical earthquake accelerations satisfy certain functional relationships with time and depth, which is applicable to the earthquake resistant design of retaining walls [19, 20]. However, the pseudo-dynamic approach also has shortcomings. For instance, seismic waves cannot meet the requirements of zero shear stress on the surface; the influence of material damping on seismic force is neglected [21, 22].