There has been much new traffic engineering construction on the loess plateau, where tunnels can be damaged by collapses based on the stability of the tunnel. Although there are currently two methods, including the laboratory test and the in situ large-scale pit immersion test, to evaluate loess collapsibility, the laboratory test method is inaccurate, and the in situ large-scale pit immersion test method is time-consuming and costly. An in situ sand well water seepage saturated loess test method was proposed for evaluating the collapsible deformation of loess strata. Four immersion tests of sand wells with different depths were carried out in a loess site on the Baoji-Lanzhou high-speed railway. Although loess strata around all sand wells at the same site are typically self-weight collapsible, the self-weight collapsible settlements of sand wells are significantly different. The settlement of the ground surface and well bottom around sand wells was measured by sand well seeping water saturated loess. Combined with the water content measurements of loess layers below and above the sand well bottom, the collapsible deformation characteristics of loess strata around sand wells were analysed to determine the self-weight collapsibility coefficient in different depth loess layers. Fissures in the ground subsidence caused by the settlement difference appeared around the sand wells. Due to the unevenness of the soil layer in the same loess site, the loess collapsibility was obviously different in adjacent sand well sites. With reference to the correction factor of self-weight collapsible deformation in the Long-xi region in China, the rationality of sand well immersion tests was preliminarily demonstrated by comparing the collapse deformation measured by the sand well tests with the value calculated by the laboratory test. The new method provides a simple and feasible means by which to quickly evaluate site and foundation collapsibility in loess tunnel engineering, which is beneficial for perfecting loess collapsibility evaluation methods and proposing measures of foundation treatment in loess tunnels.

黄土高原上出现了许多新的交通工程建设，隧道的稳定性可能会导致隧道坍塌。虽然目前评价黄土湿陷性的方法有室内试验和原位大型坑浸试验两种方法，但室内试验方法不准确，原位大型坑浸试验方法费时、费力。昂贵。提出了一种原位砂井渗水饱和黄土试验方法，用于评价黄土地层的湿陷变形。在宝兰高铁黄土场地进行了4次不同深度的砂井浸没试验。尽管同一地点所有砂井周围的黄土层通常是自重可塌陷的，砂井自重塌陷沉降差异显着。砂井周围地表和井底沉降采用砂井渗水饱和黄土测量。结合砂井底上下黄土层含水率测量，分析砂井周围黄土地层湿陷变形特征，确定不同深度黄土层的自重湿陷系数。砂井周围出现沉降差引起的地面沉降裂缝。由于同一黄土场地土层不均匀，相邻砂井场地的黄土湿陷性存在明显差异。参考我国陇西地区自重塌陷变形修正系数，通过砂井试验测得的坍塌变形与室内试验计算值的对比，初步论证了砂井浸水试验的合理性。新方法为黄土隧道工程场地及地基湿陷性快速评价提供了一种简单可行的手段，有利于完善黄土隧道工程中黄土湿陷性评价方法和提出地基处理措施。