开展陌生区域地面力学的远程、广域、快速勘测以评估车辆的越野机动性，在民用和国防等研究领域提出了迫切的需求。但目前现有的天基遥感、空基物探等技术手段属于非接触、原位勘测，只能反演部分地质信息，无法准确、快速获取地面力学参数。中科院力学所创新性提出了自由落体式贯入触探的技术手段，以突破陌生区域地面力学远程、广域、快速勘测的技术瓶颈。

自由落体式贯入触探的基本原理是贯入触探仪（贯入仪）在重力作用下获得贯入速度和动能，在贯入过程中测量贯入阻力和加速度随深度/时间的变化数据，但此测量值并不能直接获得土壤的物理力学参数，需根据贯入阻力来建立对应的参数反演方法。然而，贯入阻力与土壤力学参数的物理关系，特别是Mohr-Coulomb强度参数（黏聚力、内摩擦角）间的关系还不十分清楚。此外，当贯入速度较高时，土壤中会出现弹孔扩张现象，贯入仪与土壤的接触面积不断变化，导致了理论计算的困难。以上复杂因素导致现有的参数反演方法大都存在经验依赖性强、物理图像不清晰等问题。为此，本文针对自由落体式贯入触探开展了数值模拟及理论分析工作，对自由落体贯入过程中的贯入阻力特性及Mohr-Coulomb强度参数的反演进行了研究。

本文首先基于耦合欧拉-拉格朗日算法建立了描述自由落体贯入过程的大变形有限元分析模型，得到了贯入过程中土体应力、应变及速度场的分布规律，验证了不同初始贯入速度下土壤中存在的弹孔扩张现象。仿真结果表明，贯入过程中存在一个速度阈值，当贯入速度低于速度阈值时，贯入阻力中的静态阻力占主导地位；当贯入速度超过速度阈值时，贯入阻力中存在比较明显的动态阻力。对于给定的贯入仪-靶材组合，本文给出了一种利用数值模拟计算结果确定速度阈值范围的方法。进一步的参数分析表明：静态阻力关于土壤泊松比、弹性模量、Mohr-Coulomb强度参数以及贯入仪-土壤接触界面摩擦系数单调递增，而与土壤密度无关。

其次，本文改进了Feldgun等人提出的动态空腔膨胀模型，基于改进后的模型从理论上分析了贯入过程中的弹孔扩张现象，并计算了速度阈值的具体数值。建立了贯入仪的运动控制微分方程，提出了描述贯入过程的分阶段模型与简化模型。模型预测结果与数值模拟的计算结果符合得较好，以仿真得到的最大贯入深度为基准，分阶段模型的预测误差均小于10%，最小误差仅为2.26%。通过引入形状因子解释了理论模型与数值模拟之间静态阻力存在的偏差。

最后，本文利用自由落体式贯入触探获得的阻力-速度曲线、加速度时程曲线，基于改进后的动态空腔膨胀模型对土壤的Mohr-Coulomb强度参数建立了新的反演方法，该方法解决了现有半经验反演方法中经验依赖性强、物理图像不清晰的问题，为复杂地质环境下土壤Mohr-Coulomb参数的快速确定及土壤承载力评估提供了新的解决方案。

There is an urgent need in both civilian and defense research fields to conduct remote, wide-area, and rapid surveys of ground mechanics in unfamiliar areas to evaluate the off-road mobility of vehicles. However, the current existing remote sensing technologies such as satellite-based remote sensing and airborne geophysical exploration are non-contact and in-situ surveys, which can only invert partial geological information and cannot accurately and quickly obtain ground mechanical parameters. The Institute of Mechanics of the Chinese Academy of Sciences has innovatively proposed a free fall penetration probing technique to break through the technical bottleneck of remote, wide-area, and rapid surveys of ground mechanics in unfamiliar areas.

The fundamental principle of the free fall penetration probing technique is that the penetrating instrument (penetrometer) obtains the penetration velocity and kinetic energy under the influence of gravity. During the penetration process, it measures the variation of penetration resistance and acceleration with depth/time. However, these measurement values cannot directly obtain the physical and mechanical parameters of the soil, and require the establishment of corresponding parameter inversion methods based on the penetration resistance. However, the physical relationship between the penetration resistance and soil mechanical parameters, especially the Mohr-Coulomb strength parameters (cohesion and internal friction angle), is not fully understood. In addition, when the penetration velocity is high, the cavitation phenomenon may occur in the soil, which causes the contact area between the penetrometer and the soil to constantly change, leading to difficulties in theoretical calculations. These complex factors have resulted in problems such as strong empirical dependence and unclear physical images in existing parameter inversion methods. Therefore, this paper conducted numerical simulations and theoretical analyses on the free-fall penetration probing technique and studied the characteristics of penetration resistance and the inversion of Mohr-Coulomb strength parameters during the free-fall penetration process.

Firstly, this paper used a coupled Eulerian-Lagrangian algorithm to establish a large deformation finite element analysis model that describes the free-fall penetration process. The distribution laws of soil stress, strain, and velocity fields during the penetration process were obtained, and the cavitation phenomenon existing in the soil at different initial penetration velocities was verified. Simulation results indicate that there exists a critical velocity during the penetration process. When the penetration velocity is lower than the critical velocity, the static resistance dominates the penetration resistance; when the penetration velocity exceeds the critical velocity, there is a significant dynamic resistance in the penetration resistance. For a given penetrometer-target combination, this paper proposes a method for determining the range of critical velocities using numerical simulation results. Further parameter analysis shows that the static resistance monotonically increases with soil Poisson's ratio, elastic modulus, Mohr-Coulomb strength parameters, and friction coefficient of the penetrometer-soil contact interface but is independent of soil density.

Secondly, this paper improves the dynamic cavity expansion model proposed by Feldgun et al. Based on the improved model, the cavitation phenomenon during the penetration process is theoretically analyzed, and the specific value of the critical velocity is calculated. The motion control differential equation of the penetrometer is established, and a staged model and simplified model are proposed to describe the penetration process. The predicted results of the models match well with the numerical simulation results. Using the maximum penetration depth obtained from simulations as a benchmark, the prediction error of the staged model is less than 10%, and the minimum error is only 2.26%. By introducing a shape factor, the deviation in the static resistance between the theoretical model and numerical simulation is explained.

Finally, this paper proposes a new inversion method for the soil Mohr-Coulomb strength parameters based on the improved dynamic cavity expansion model using the resistance-velocity curve and acceleration-time curve obtained from the free fall penetration test. This method solves the problem of strong empirical dependence and unclear physical image in existing semi-empirical inversion methods, providing a new solution for rapid determination of soil Mohr-Coulomb parameters and soil bearing capacity evaluation in complex geological environments.