水力压裂过程的扩展有限元数值模拟方法

建立了基于扩展有限元法的水力压裂数值模拟方法,使水力裂缝独立于网格存在,无需预设裂缝扩展方位。在扩展有限元计算框架下,将裂缝面处理为求解域内边界,将缝内水压力转化为相关单元等效节点力;运用考虑缝内水压力作用的相互积分法来数值求解缝尖应力强度因子;采用最大能量释放率准则确定裂缝是否继续扩展及扩展方位;最终编制了计算机程序。利用该方法数值模拟了单条水力裂缝在恒定水压力作用下作非平面扩展,所得结果分别与室内试验和解析模型相对比。结果表明,数值结果与室内试验和解析解吻合较好,缝尖应力强度因子最大相对误差不高于0.45%,验证了该方法的可行性和准确性。     

基于动力显式算法的热成形数值模拟预测

考虑金属热处理相变动力学模型,建立了适用于高强度钢板热成形的热、力、相变多场耦合本构方程。基于大变形动力显式有限元算法及上述多场耦合本构方程,建立了热成形动力显式有限元方程。并将板料相变潜热释放引入到热成形温度场分析过程中。在自主开发的商业化金属成形CAE软件KMAS(King Mesh Analysis System)基础上开发了热成形动力显式分析模块,可用于预测热成形过程中零件的厚度变化、温度变化、微观组织各相的体积分数以及硬度分布。随后采用该模块对一款汽车B柱的热成形过程及最终力学性能进行数值模拟预测,并与试验结果进行对比,对比的一致性证明了所建立的多场耦合本构方程及KMAS热成形动力显式分析模块的正确性。     

基于神经网络的白酒勾兑目标规划算法优化

针对白酒勾兑过程中,现有的白酒勾兑目标规划算法难以确定权系数(优先因子)的缺点,本文提出了利用人工神经网络对目标规划算法进行改进和优化,选择三层前向BP神经网络结构,并通过选取理化指标向量与\     

“十”字地铁换乘站多点火灾工况下的通风排烟

以某典型“十”字交叉换乘地铁车站为研究对象, 通过数值模拟的方法对换乘车站两站台同时发生火灾的极端情况下不同排烟方案的排烟效果进行了研究, 方案1是站厅层送风, 地下二层和三层站台分别排烟;方案2是在方案1的基础上, 将地下三层的车站屏蔽门打开, 利用车站隧道风机进行辅助排烟.火源设计为峰值2 MW的时间平方快速火.研究结果表明:两种方案都能有效控制地下二层站台烟气蔓延, 方案2中打开地下三层车站屏蔽门, 利用车站隧道系统辅助排烟的措施可以有效增加楼梯处的向下风速, 同时降低楼梯处烟气温度, 延长火源附近楼梯作为人员疏散通道的时间, 实现良好的换乘站防烟效果.     

沥青路面铺筑温度场数值仿真模型

为确定一定环境下沥青混合料铺筑过程温度衰减规律, 对铺筑过程路面温度场的仿真模型进行了研究.采用非连续的随碾压次数而变的层厚、材料热物性和洒水散热条件, 连续的随温度而变的对流换热、辐射换热条件, 以及非均匀的底层初始温度场, 建立了一维仿真模型.通过变换几何模型及循环调用初始温度场的方法, 实现了非连续条件在连续场分析中的应用.由仿真模型验证, 本文模型可以在不同环境条件下实现铺筑过程温度场仿真.     

Temperature Rise Inside Shear Bands in a Simple Model Glass

One of the key factors, which hampers the application of metallic glasses as structural components, is the localization of deformation in narrow bands of a few tens up to one hundred nanometers thickness, the so-called shear bands. Processes, which occur inside shear bands are of central importance for the question whether a catastrophic failure of the material is unavoidable or can be circumvented or, at least, delayed. Via molecular dynamics simulations, this study addresses one of these processes, namely the local temperature rise due to viscous heat generation. The major contribution to energy dissipation is traced back to the plastic work performed by shear stress during steady deformation. Zones of largest strain contribute the most to this process and coincide with high-temperature domains (hottest spots) inside the sample. Magnitude of temperature rise can reach a few percent of the sample’s glass transition temperature. Consequences of these observations are discussed in the context of the current research in the field.     

The Lattice Model of Particles with Orientation-Dependent Interactions at Solid Surfaces: Wetting Scenarios

Wetting phenomena in a lattice model of particles having two chemically different halves (A and B) and being in contact with solid substrates have been studied with Monte Carlo methods. The energy of the interaction between a pair of neighboring particles has been assumed to depend on the degree to which the AA, AB and BB regions face each other. In this work, we have assumed that uAA=−1.0 and considered three series of systems with uAB=uBB, uAB=0 and uBB=0. The phase behavior of bulk systems has been determined. In particular, it has been shown that at sufficiently low temperatures the bulk systems order into the superantiferromagnetic (SAF) phase, or into the antiferromagnetic (AF) phase, depending on the magnitudes of AA, AB and BB interaction energies, uAA, uAB and uBB. The SAF structure occurs whenever ϵ=uAA+uBB−2uAB is lower than zero and the AF structure is stable when ϵ is greater than zero. The wetting behavior has been demonstrated to depend strongly on the structure of the bulk condensed phase, the interactions between fluid particles and the strength of the surface potential. In all series, we have found the dewetting transition, resulting from the limited stability of different ordered structures of surface phases. However, in the systems that exhibit the gas–liquid transition in the bulk, the reentrant wetting transition has been observed at sufficiently high temperatures. The mechanism of dewetting and reentrant wetting transitions has been determined. Moreover, we have also demonstrated, how the dewetting transition in the series with uAB=0 is affected by the wall selectivity, i.e., when the interaction between the parts A and B of fluid particles and the solid is different.     

State-of-the-Art Molecular Dynamics Simulation Studies of RNA-Dependent RNA Polymerase of SARS-CoV-2

Molecular dynamics (MD) simulations are powerful theoretical methods that can reveal biomolecular properties, such as structure, fluctuations, and ligand binding, at the level of atomic detail. In this review article, recent MD simulation studies on these biomolecular properties of the RNA-dependent RNA polymerase (RdRp), which is a multidomain protein, of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are presented. Although the tertiary structures of RdRps in SARS-CoV-2 and SARS-CoV are almost identical, the RNA synthesis activity of RdRp of SARS-CoV is higher than SARS-CoV-2. Recent MD simulations observed a difference in the dynamic properties of the two RdRps, which may cause activity differences. RdRp is also a drug target for Coronavirus disease 2019 (COVID-19). Nucleotide analogs, such as remdesivir and favipiravir, are considered to be taken up by RdRp and inhibit RNA replication. Recent MD simulations revealed the recognition mechanism of RdRp for these drug molecules and adenosine triphosphate (ATP). The ligand-recognition ability of RdRp decreases in the order of remdesivir, favipiravir, and ATP. As a typical recognition process, it was found that several lysine residues of RdRp transfer these ligand molecules to the binding site such as a “bucket brigade.” This finding will contribute to understanding the mechanism of the efficient ligand recognition by RdRp. In addition, various simulation studies on the complexes of SARS-CoV-2 RdRp with several nucleotide analogs are reviewed, and the molecular mechanisms by which these compounds inhibit the function of RdRp are discussed. The simulation studies presented in this review will provide useful insights into how nucleotide analogs are recognized by RdRp and inhibit the RNA replication.