为解决土压平衡盾构施工中渣土物理力学性质检测依赖人工取样、周期长、难以及时指导施工的问题，提出一种适用于盾构施工现场的渣土状态自动化检测方法。研制基于十字板剪切试验原理，集自动取样、剪切检测、数据处理及自动清土于一体的自动化检测设备；
基于已有理论关系，构建渣土剪切强度与坍落度之间的换算方法，并开发自动化控制程序，实现渣土剪切强度、坍落度、温度及有害气体浓度等参数的在线快速检测。依托天津地铁8号线鞍山西道站—南丰路站区间盾构工程开展现场应用试验，完成设备调试及自动化流程验证，对比分析不同工况下渣土状态参数与盾构刀盘转矩、总推力等掘进参数的响应关系。结果表明： 1）设备可实现连续掘进过程中的自动取样与在线检测，单次检测全流程耗时410～420 s（其中剪切检测耗时约20
s）； 2）现场12组代表性工况检测剔除剪切板启停阶段数据后，稳定阶段数据均值处理能够有效降低检测波动，获得稳定的检测结果； 3）渣土坍落度与盾构掘进参数具有良好的对应关系，随着坍落度增大，刀盘转矩与总推力整体呈下降趋势，且刀盘转矩对坍落度的变化更为敏感。研究表明，该自动化检测设备能有效评价渣土改良效果并反映掘进负载变化，可为盾构施工参数的动态优化提供参考。
Physicomechanical property testing of muck generated from earth pressure balance shield tunneling relies heavily on manual sampling with long cycles and fails to timely reflect construction conditions. To address these challenges, an automated muck state detection method for shield construction sites is proposed. Based on the vane shear test principle, an automated detection device integrating automatic sampling, shear testing, data processing, and automatic muck cleaning is developed. A conversion method between muck shear strength and slump is established based on existing theoretical relationships, and an automated control program is developed for rapid online detection of parameters, including muck shear strength, slump, temperature, and hazardous gas concentration. Field tests were conducted in the shield section between the Anshanxidao Station and the Nanfenglu Station of the Tianjin Metro Line 8. Equipment commissioning and automated process verification were completed, and the relationships between muck state and shield tunneling parameters, such as cutterhead torque and total thrust, were comparatively analyzed under different working conditions. The results are as follows: (1) The device enables automatic sampling and online detection during continuous tunneling, with a single detection cycle of 410－420 s (approximately 20 s for the shear test). (2) Detection of 12 groups of representative field working conditions shows that removing the data from the vane shear plate start and stop stages and averaging data from the stable stage can effectively reduce detection fluctuations and yield stable results. (3) The detected muck slump correlates well with shield tunneling parameters. As slump increases, cutterhead torque and total thrust generally decrease, with cutterhead torque more sensitive to slump changes. The study indicates that the proposed automated detection device can effectively evaluate the muck conditioning effect and reflect tunneling load variation, providing a reference for the dynamic optimization of shield construction parameters.
