Simple lattice Boltzmann approach for turbulent buoyant flow simulation

In the present work, a simple large eddy simulation （LES）-based lattice Boltz- mann model （LBM） is developed for thermal turbulence research. This model is validated by some benchmark tests. The numerical results demonstrate the good performance of the present model for turbulent buoyant flow simulation.     

Lattice Boltzmann simulations of turbulent shear flow between parallel porous walls

The effects of two parallel porous walls are investigated, consisting of the Darcy number and the porosity of a porous medium, on the behavior of turbulent shear flows as well as skin-friction drag. The turbulent channel flow with a porous surface is directly simulated by the lattice Boltzmann method （LBM）. The Darcy-Brinkman- Forcheimer （DBF） acting force term is added in the lattice Boltzmann equation to simu- late the turbulent flow bounded by porous walls. It is found that there are two opposite trends （enhancement or reduction） for the porous medium to modify the intensities of the velocity fluctuations and the Reynolds stresses in the near wall region. The parametric study shows that flow modification depends on the Darcy number and the porosity of the porous medium. The results show that, with respect to the conventional impermeable wall, the degree of turbulence modification does not depend on any simple set of param- eters obviously. Moreover, the drag in porous wall-bounded turbulent flow decreases if the Darcy number is smaller than the order of O（10-4） and the porosity of porous walls is up to 0.4.     

基于支持向量机的结构损伤分步识别研究

支持向量机是一种基于统计学习理论的机器学习算法,能够较好地解决小样本的学习问题.本文介绍了支持向量机分类和回归算法,提出了基于支持向量机的结构损伤分步识别方法:以模态频率作为损伤特征,首先根据支持向量机分类算法的概率估计确定可能的损伤位置,重新构造训练样本,然后利用支持向量机回归算法计算损伤位置;最后估计损伤程度.以梁的损伤识别为例进行了验证,结果表明该方法可以提高损伤识别的精度.     

基于CCH的SVM几何算法及其应用

The support vector machine （SVM） is a novel machine learning tool in data mining. In this paper, the geometric approach based on the compressed convex hull （CCH） with a mathematical framework is introduced to solve SVM classification problems. Compared with the reduced convex hull （RCH）, CCH preserves the shape of geometric solids for data sets; meanwhile, it is easy to give the necessary and sufficient condition for determining its extreme points. As practical applications of CCH, spare and probabilistic speed-up geometric algorithms are developed. Results of numerical experiments show that the proposed algorithms can reduce kernel calculations and display nice performances.     

On the interaction between two fixed spherical particles

The variation of the drag (C_D) and lift coefficients (C_L) of two fixed solid spherical particles placed at different positions relative each other is studied. Simulations are carried out for particle Reynolds numbers of 50, 100 and 200 and the particle position is defined by the angle between the line connecting the centers of the particles and the free-stream direction (α) and the separation distance (d₀) between the particles. The flow around the particles is simulated using two different methods; the Lattice Boltzmann Method (LBM), using two different computational codes, and a conventional finite difference approach, where the Volume of Solid Method (VOS) is used to represent the particles. Comparisons with available numerical and experimental data show that both methods can be used to accurately resolve the flow field around particles and calculate the forces the particles are subjected to. Independent of the Reynolds number, the largest change in drag, as compared to the single particle case, occurs for particles placed in tandem formation. Compared to a single particle, the drag reduction for the secondary particle in tandem arrangement is as high as 60%, 70% and 80% for Re = 50, 100 and 200, respectively. The development of the recirculation zone is found to have a significant influence on the drag force. Depending on the flow situation in-between the particles for various particle arrangements, attraction and repulsion forces are detected due to low and high pressure regions, respectively. The results show that the inter-particle forces are not negligible even under very dilute conditions.     

数据驱动下人体面部皮肤的本构参数反演及力学特征分析

人体面部皮肤的本构参数反演及力学特性研究对皮肤病变早期诊断、皮肤仿生材料设计及计算图形学中的面部模型建立都有着至关重要的作用, 机器学习与有限元仿真方法相结合能更高效、更准确地解决非入侵式皮肤组织本构参数反求问题. 首先建立了面部皮肤多向拉伸下的应力松弛有限元模型, 并通过多步位移控制法对皮肤的超弹性力学性能与黏弹性力学性能分离; 对Gasser-Ogden-Holzapfel (GOH)与Prony series本构模型参数进行敏感性分析, 揭示影响面部皮肤应力松弛实验结果的关键参数. 其次, 利用贝叶斯超参数优化理论搭建随机森林(RF)模型与支持向量回归(SVR)模型, 结合实验数据对人体面部皮肤组织本构参数进行了反求. 最后, 将计算得到的有限元仿真曲线与试验获得的拉伸力响应曲线对比, 并引入决定系数 R²对两种模型的预测准确性进行了评估. 结果表明, 纤维组织分散度κ、剪切模量相关参数C₁₀和松弛模量g₁是影响皮肤应力松弛实验结果的关键参数, RF模型数值计算曲线与试验曲线的拟合优度为0.98, 其在皮肤本构参数反演问题上表现出更高的准确率, 机器学习可以精准高效地获取面部皮肤的本构参数, 进而准确描述皮肤组织的力学性能, 该方法也可进一步推广到其他生物软组织的复杂本构参数反演问题.     

Free convection of nanofluid filled enclosure using lattice Boltzmann method （LBM）

The lattice Boltzmann method (LBM) is used to examine free convection of nanofluids. The space between the cold outer square and heated inner circular cylinders is filled with water including various kinds of nanoparticles: TiO₂, Ag, Cu, and Al₂O₃. The Brinkman and Maxwell-Garnetts models are used to simulate the viscosity and the effective thermal conductivity of nanofluids, respectively. Results from the performed numerical analysis show good agreement with those obtained from other numerical methods. A variety of the Rayleigh number, the nanoparticle volume fraction, and the aspect ratio are examined. According to the results, choosing copper as the nanoparticle leads to obtaining the highest enhancement for this problem. The results also indicate that the maximum value of enhancement occurs at λ = 2.5 when Ra = 10⁶ while at λ = 1.5 for other Rayleigh numbers.     

Lattice Boltzmann method for simulating particle–fluid interactions

The lattice Boltzmann method （LBM） has gained increasing popularity in the last two decades as an alternative numerical approach for solving fluid flow problems. One of the most active research areas in the LBM is its application in particle-fluid systems, where the advantage of the LBM in efficiency and parallel scalability has made it superior to many other direct numerical simulation （DNS） techniques. This article intends to provide a brief review of the application of the LBM in particle-fluid systems. The numerical techniques in the LBM pertaining to simulations of particles are discussed, with emphasis on the advanced treatment for boundary conditions on the particle-fluid interface. Other numerical issues, such as the effect of the internal fluid, are also briefly described. Additionally, recent efforts in using the LBM to obtain closures for particle-fluid drag force are also reviewed.