Shihai Li

A Comparative Study on Different Parallel Solvers for Nonlinear Analysis of Complex Structures

The parallelization of 2D/3D software SAPTIS is discussed for nonlinear analysis of complex structures. A comparative study is made on different parallel solvers. The numerical models are presented, including hydration models, water cooling models, modulus models, creep model, and autogenous deformation models. A finite element simulation is made for the whole process of excavation and pouring of dams using these models. The numerical results show a good agreement with the measured ones. To achieve a better computing efficiency, four parallel solvers utilizing parallelization techniques are employed: (1) a parallel preconditioned conjugate gradient (PCG) solver based on OpenMP, (2) a parallel preconditioned Krylov subspace solver based on MPI, (3) a parallel sparse equation solver based on OpenMP, and (4) a parallel GPU equation solver. The parallel solvers run either in a shared memory environment OpenMP or in a distributed memory environment MPI. A comparative study on these parallel solvers is made, and the results show that the parallelization makes SAPTIS more efficient, powerful, and adaptable.

A GPU-Based Parallel Procedure for Nonlinear Analysis of Complex Structures Using a Coupled FEM/DEM Approach

This study reports the GPU parallelization of complex three-dimensional software for nonlinear analysis of concrete structures. It focuses on coupled thermomechanical analysis of complex structures. A coupled FEM/DEM approach (CDEM) is given from a fundamental theoretical viewpoint. As the modeling of a large structure by means of FEM/DEM may lead to prohibitive computation times, a parallelization strategy is required. With the substantial development of computer science, a GPU-based parallel procedure is implemented. A comparative study between the GPU and CPU computation results is presented, and the runtimes and speedups are analyzed. The results show that dramatic performance improvements are gained from GPU parallelization.

Real-time monitoring instrument designed for the deformation and sliding period of colluvial landslides

A real-time monitoring instrument for colluvial landslides that can capture horizontal displacement during the deformation period of the slope, deep dislocation displacement, accurate depth of the slip surface, and sliding direction during the initial sliding period has been developed. The horizontal displacement monitoring system is a monitoring apparatus assembled in an array based on the MEMS (Micro-Electro-Mechanical System). Horizontal displacement is obtained by integration of the X- and Y-direction angles along the drilling direction, and the sliding direction is simultaneously measured via vector composition of the X- and Y-direction angles obtained from the MEMS sensor. Deep dislocation displacement is calculated based on the relative displacement between the base point embedded in the bedrock and measuring point above the sliding surface, and measurement of relative displacement is executed by an angular displacement sensor fixed on top of the inclinometer tube. The accurate depth measurement of the slip surface is transferred to the judgment about the on-off state of the signal wire. Based on the long-term monitoring of the Yu Jiaba landslide at the Three George dams, an open-pit mine in Inner Mongolia province, and an Anshan open-pit mine, the results reveal that the reliability of the instrument is very high. High accuracy, ability to capture the entire landslide process, large deformation monitoring, autonomous data collection, and reusable characteristics could be implemented.

The Three-Stage Model Based on Strain Strength Distribution for the Tensile Failure Process of Rock and Concrete Materials

A three-stage model is introduced to describe the tensile failure process of rock and concrete materials. Failure of the material is defined to contain three stages in the model, which include elastic deformation stage, body damage stage and localization damage stage. The failure mode change from uniform body damage to localization damage is expressed. The heterogeneity of material is described with strain strength distribution. The fracture factor and intact factor, defined as the distribution function of strain strength, are used to express the fracture state in the failure process. And the distributive parameters can be determined through the experimental stress-strain curve.

2D particle contact-based meshfree method in CDEM and its application in geotechnical problems

Purpose - Continuum-based discrete element method is an explicit numerical method, which is a combination of block discrete element method (DEM) and FEM. When simulating large deformation problems, such as cutting, blasting, water-like material flowing, the distortion of elements will lead to no convergence of the numerical system. To solve the convergence problem, a particle contact-based meshfree method (PCMM) is introduced in. The paper aims to discuss this issue. Design/methodology/approach - PCMM is based on traditional particle DEM, and use particle contacts to generate triangular elements. If three particles are contact with each other, the element will be created. Once elements are created, the macroscopic constitutive law could be introduced in. When large deformation of element occurs, the contact relationship between particles will be changed. Those elements that do not meet the contact condition will be deleted, and new elements that coincide with the relationship will be generated. By the deletion and creation of elements, the convergence problem induced by element distortion will be eliminated. To solve FEM and PCMM coupled problems, a point-edge contact model is introduced in, and normal and tangential springs are adopted to transfer the contact force between particles and blocks. Findings - According to the deletion and recreation of elements based on particle contacts, PCMM could simulate large deformation problems. Some numerical cases (i.e. elastic field testing, uniaxial compression analysis and wave propagation simulation) show the accuracy of PCMM, and others (i.e. soil cutting, contact burst and water-like material flowing) show the rationality of PCMM. Originality/value - In traditional particle DEM, contact relationships are used to calculate contact forces. But in PCMM, contact relationships are adopted to generate elements. Compared to other meshfree methods, in PCMM, the element automatic deletion and recreation technique is used to solve large deformation problems.