为探明盾构地中对接过程中水泥改良粉质黏土地层的温度场演化特征及其冻胀抑制机制，以江阴靖江长江隧道盾构水下对接工程为原型，基于相似准则开展大型3向加载物理模型试验，系统对比分析原状粉质黏土与质量掺量为10%的水泥改良粉质黏土在非对称冻结形式下的温度场演变规律、冻结壁发展过程及冻胀力时空分布特征，并对冻胀抑制效果进行定量评价。试验结果表明： 1）水泥改良后地层降温速率降低，冻结壁交圈时间较原状土延长约22.7%，但交圈后冻结锋面发展速率加快； 冻结90 d后，2类土层冻结壁厚度差异显著缩小。2）冻结过程中，冻胀力呈阶段性演化特征，原状土在冻结壁交圈前后冻胀力增长显著，而水泥改良土冻胀力发展存在明显滞后性，冻胀力峰值出现时间延后。3）最大冻胀荷载集中于盾构对接截面腰线位置，原状土最大冻胀力可达地层初始压力的2倍以上，经质量掺量为10%的水泥改良后，同位置最大冻胀力降至1.48倍，整体冻胀力平均降低约40%，对接中心区域抑制效果最为显著，最大降幅达52%。4）水泥改良通过改变土体的热物理性质及水分迁移特征，进而影响冻结过程中温度场与冻胀力发展规律，工程中可通过合理设计水泥改良措施及掺量调节冻结壁形成过程及冻胀演化特征。
To investigate the temperature-field evolution and frost-heave suppression mechanism of cement-improved silty clay during underground shield docking, a case study is conducted on the underwater shield docking project of the Jiangyin-Jingjiang Yangtze River Tunnel. A large-scale three-dimensional loading physical model is constructed based on similarity criteria, and a comparative analysis is conducted on the temperature-field evolution, frozen-wall development, and spatiotemporal distribution of frost heave force between undisturbed silty clay and silty clay improved with 10% cement under asymmetric freezing conditions. Furthermore, the frost-heave suppression effect is quantitatively evaluated. The results show that the cement-improved ground has a lower cooling rate and a ~22.7% longer frozen-wall closure time than undisturbed soil. However, the development rate of the freezing front accelerates after closure, resulting in a significant reduction in the thickness difference between the frozen walls of the two soils after 90 days. Furthermore, the frost heave force exhibits a staged evolution during freezing: it increases significantly in the undisturbed soil before and after frozen-wall closure, whereas a noticeable lag is observed in the cement-improved soil, with a delayed peak occurrence. The maximum frost heave load is concentrated at the springline of the docking section. Specifically, the peak frost heave force in the undisturbed soil is more than double the initial ground pressure, whereas it decreases to 1.48 times the initial ground pressure after cement improvement. This corresponds to an average reduction of ~40% in the overall frost heave force, with a maximum suppression rate of 52% achieved in the central docking area. The cement improvement alters the thermophysical properties and moisture migration characteristics of the soil, influencing the evolution of the temperature field and frost heave force during freezing. Consequently, the frozen-wall formation process and frost heave response can be effectively regulated by appropriately designing cement improvement measures and dosages.
