提出了一种颗粒形状影响下碎石地层中碎石-桩界面侧摩阻力计算方法,使用可描述应力应变硬化及软化的界面剪切函数计算单桩侧摩阻力;为探究碎石颗粒形状对界面剪切应力应变行为的影响,制备了八种碎石,使用图像处理的方法获得碎石的形状参数(长细比、凸度及球度);考虑三种形状参数的权重,提出了一种改进的颗粒整体规整度的计算方法;建立不规则碎石-桩界面剪切的离散元模型,探究了颗粒整体规整度与临界界面摩擦角、界面剪切临界状态之间的关系.结果表明:颗粒整体规整度与临界界面摩擦角之间存在很强的线性关系,且整体规整度只改变临界状态线的截距.在桩侧摩阻函数中考虑颗粒形状的影响,并求解碎石地层中竖向荷载下不同深度桩基础桩侧摩阻力.将室内试验结果与采用本文提出的考虑颗粒形状的桩侧摩阻计算方法获得结果进行对比,验证了本文所提出方法的适用性及准确性.

The Direct Simulation Monte Carlo (DSMC) method is the most widely used technique for studying complex non-equilibrium rarefied gas flows, and the DSMC method based on unstructured meshes is easy to handle complex geometries, but has higher requirements for computing resources. In this paper, a large-scale parallel acceleration study is carried out on homogeneous and heterogeneous computer architectures by analyzing the data structure, numerical methods and computational processes basing on an unstructured DSMC solver. For the homogeneous architecture, in order to minimize data communication and maximize parallelism, an efficient data communication method and an MPI+OpenMP hybrid parallel algorithm are designed, and several performance optimization methods are discussed. For the heterogeneous architecture, in order to alleviate the complexity of GPU memory management and improve data locality, the MPI+OpenACC hybrid parallel strategy based on unified memory technology is proposed. Using a 3D hypersonic plate flow, the accuracy of the parallel algorithm is verified. For large-scale test studies, the MPI+OpenMP hybrid parallel strategy achieves an acceleration ratio of 6.18 compared to the pure MPI parallel strategy at the 8,000-core computing scale. At the same time, 60,000-core hybrid parallel computing has obtained an acceleration ratio of 5.54 (compared to 1024 cores), and 8-GPU parallel computing has obtained an acceleration ratio of 32.04 (compared to CPU single core), which has strong parallel acceleration capabilities and good scalability. In addition, the portability on homogeneous and heterogeneous computing platforms is also compared and analyzed, which provides reference for parallel computing of similar application problems.

An image-based gradation calculation method considering crushed stone morphology is proposed in this paper. Eight kinds of crushed stones are prepared to figure out the effect of the particle shape on the gradation. All particles are pictured, and the photos are processed to obtain particle shape indices by using the image-based method. A minimum Feret diameter analysis method is used, which can reflect the influence of the particle shape. Then, the paper proposes a developed overall regularity to describe the morphology of particles, which is calculated by the weight of three shape indices: aspect ratio, convexity, and sphericity. The developed overall regularity is applied on gradation curve equations to provide an image-based gradation calculation method. And the provided method is validated to well fit the actual case. Furthermore, the method is also applied on two examples of gradation designation and the result proves to be feasible and practicable.

A new method is proposed to calibrate the key parameters of a two-surface elastoplastic model to simulate the sand-structure interface under monotonic shear loading. A series of DEM simulations, in which a few sandstructure interface parameters and boundary conditions are considered, are conducted to guide the establishment of a calibration method. The simulation results, together with other researchers? test results, are used to provide semiempirical and empirical equations for parameter calibration using the two-surface elastoplastic constitutive model. Numerical calculations are performed using the new method and the model?s performance is evaluated for dense and loose sands, under constant normal load and constant normal stiffness conditions by comparison with experimental results, which are shown to be in good agreement with the experimental data.