基于HBP本构模型的泥石流动力过程SPH数值模拟

本构模型是描述泥石流流变特性的关键,也是决定其动力过程数值模拟准确性的核心问题之一。泥石流流体属多相混合物,现有的研究已证实其存在剪切增稠或剪切变稀的现象,传统基于Bingham及Cross线性本构关系的数值模型难以准确描述泥石流流变特性。文中探讨了Bingham模型在低剪应变率下的数值发散问题,在光滑粒子流体动力学（SPH）方法框架上建立了整合Herschel-Bulkley-Papanastasiou(HBP）本构关系的稀性泥石流动力过程三维数值模型。相比传统基于浅水波假设的二维数值模型,所述方法从三维尺度建立SPH形式下的泥石流浆体纳维?斯托克斯方程并进行数值求解,可获取泥石流速度场时空分布及堆积形态,同时采用HBP本构关系描述泥石流流变特性,能在确保数值收敛的前提下反映泥石流流体在塑性屈服过渡段及大变形状态下应力?应变的非线性变化。为验证提出方法的合理性,结合小型模型槽实验观测进行了对比,结果表明数值模拟与实测结果基本吻合。     

蒙古气旋天气过程中的沙尘传输特征

基于沙尘数值预报模式针对不同区域地面起沙的敏感性试验结果,分析讨论了蒙古气旋沙尘暴过程中沙尘传输的特点及形成原因。结果表明：蒙古气旋沙尘暴过程的沙尘传输表现为:沙源区纬度越高,沙尘向东传输越强,纬度越低,向南传输越强;同时,高度越高,沙尘向东传输越强,高度越低,向南传输越强。其形成原因是萨彦岭山地背风坡效应、青藏高原东北侧地形强迫绕流等自然地理因素和蒙古气旋的动力、热力结构共同造成的,具有一定程度的普遍性。     

刚性桩复合地基侧向土压力试验研究

通过室内模型试验,实测得到试验条件下,天然地基和刚性桩复合地基作用在不允许有位移的刚性挡土墙上的侧向土压力;通过与天然地基对比分析,获得了刚性桩复合地基的侧向土压力特性及分布规律。结果表明,刚性桩复合地基中桩的参与（包括桩的荷载深层传递作用、桩负摩擦区的影响和桩体对桩间土水平附加应力的“遮拦”作用等）使复合地基侧向土压力大小和分布规律明显区别于天然地基;在给定荷载水平下,刚性桩复合地基的侧向土压力值低于天然地基,侧向土压力影响范围较天然地基作用位置更深;在试验条件下,刚性挡土墙距离建筑物0.35～1.4 m范围内,当荷载水平达到地基承载力特征值时,刚性桩复合地基作用在刚性挡土墙上的总土压力和附加土压力约为天然地基的43.3%～80.1%和15.9%～59.8%。     

智能优化LSSVM算法的混沌时间序列边坡变形预测模型

边坡变形序列存在一定混沌特征,本文将混沌分析方法应用到边坡变形预测中。为解决标准LSSVM模型中惩罚参数和核函数参数因任意给定或经验给定带来的非最优问题,将遗传算法和粒子群算法引入LSSVM模型,根据变形序列建立GA-LSSVM和PSO-LSSVM预测模型,与标准LSSVM混沌预测模型和基于神经网络的混沌预测模型进行比较。结果表明,GA-LSSVM和PSO-LSSVM模型预测中误差分别为0.73 mm和0.77 mm,LSSVM,BP,RBF三种模型中误差分别为0.90 mm,0.80 mm和0.75 mm;因此,本文提出的预测模型用于边坡变形预测比其他几种模型具有更高的精度。     

基于SVM的甲壳动物线粒体基因分析方法

甲壳动物线粒体基因组蕴涵了物种进化历程中重要的遗传信息,如何有效地利用这些保留在基因组中的基因序列和基因顺序信息,是甲壳动物线粒体基因组研究的一个重点方向。为了进一步探讨甲壳动物稳定、可靠的系统发育关系,本文利用支持向量机的分类功能实现了甲壳动物线粒体基因组基因区与基因间区、编码区与非编码区的准确分类和预测,同时为了提高分类学习机的泛化能力,使用了交叉验证方法和粒子群算法优化选取支持向量机相关训练参数。通过MATLAB仿真分析的方法,对10种甲壳动物线粒体基因组序列的基因区和基因间区进行分类,以及对5种甲壳动物进行线粒体基因组序列中编码区和非编码区的分类,获得了较好的分类准确率。仿真结果表明本文方法是可行的和有效的,能够出色地应用于甲壳动物线粒体基因组序列的研究分析。     

Variation of Flow Field Around Twin Cylinders with and Without the Outer Perforated Cylinder ? Numerical Study

Presence of the outer perforated cylinder reduces the direct wave impact on the inner cylinder, which has been testified by many researchers. However, the force reduction mechanism, which is complicated due to the wave-porous structure interaction, needs to be addressed in detail. The present study explains the mechanism with the aid of the computational fluid dynamics (CFD) tool STAR CCM+. This package is chosen for its capabilities to simulate viscous and turbulence effects caused by passage of waves. For the present study, flow fields around the twin cylinders with different orientations are examined with and without the outer perforated cover. Mechanism contributing to the reduction of force on the existing structure is explained in physical terms, and force reduction is quantified. The present study has direct application in the retrofitting application of offshore members.     

A second-order particle tracking method

An accurate particle tracking method for a finite difference method model is developed using a constant acceleration method. Being assumed constant temporal and spatial gradients, the new method permits temporal-spatial variability of particle velocity. Test results in a solid rotating flow show that the new method has second-order accuracy. The performance of the new method is compared with that of other methods; the first-order Euler forward method, and the second-order Euler predictorcorrector method. The new method is the most efficient method among the three. It is more accurate and efficient than the other two.     

黄海绿潮应急溯源数值模拟初步研究

基于三维全动力POM海洋模式,根据2008年6月1日海监飞机监测绿潮所在位置,采用拉格朗日粒子追踪法反向积分,追溯绿潮来源。数值模拟结果显示,回溯至5月中旬,绿潮主要来源于黄海南部江苏连云港和盐城近海海域。黄海绿潮溯源数值模拟,为政府相关部门了解绿潮的源头,并采取相应的措施提供依据,进而为保护生态环境、防灾减灾做贡献。     

Enhancement of Solar Energetic Particles During a Shock – Magnetic Cloud Interacting Complex Structure

The behavior of solar energetic particles (SEPs) in a shock – magnetic cloud interacting complex structure observed by the Advanced Composition Explorer (ACE) spacecraft on 5 November 2001 is analyzed. A strong shock causing magnetic field strength and solar wind speed increases of about 41 nT and 300 km s⁻¹, respectively, propagated within a preceding magnetic cloud (MC). It is found that an extraordinary SEP enhancement appeared at the high-energy (≥10 MeV) proton intensities and extended over and only over the entire period of the shock – MC structure passing through the spacecraft. Such SEP behavior is much different from the usual picture that the SEPs are depressed in MCs. The comparison of this event with other top SEP events of solar cycle 23 (2000 Bastille Day and 2003 Halloween events) shows that such an enhancement resulted from the effects of the shock – MC complex structure leading to the highest ≥10 MeV proton intensity of solar cycle 23. Our analysis suggests that the relatively isolated magnetic field configuration of MCs combined with an embedded strong shock could significantly enhance the SEP intensity; SEPs are accelerated by the shock and confined into the MC. Further, we find that the SEP enhancement at lower energies happened not only within the shock – MC structure but also after it, probably owing to the presence of a following MC-like structure. This is consistent with the picture that SEP fluxes could be enhanced in the magnetic topology between two MCs, which was proposed based on numerical simulations by Kallenrode and Cliver (Proc. 27th ICRC 8, 3318, 2001b).     

基于仿生原理的粒子群算法求解江安校区水动力弥散系数

水动力弥散系数是研究地下水溶质运移的一个重要参数。为了解污染物在地下水中的运移规律,利用基于仿生学原理的粒子群算法,求解四川大学江安校区弥散试验场中的潜水含水层天然流场下的水动力弥散系数,并与最小二乘法和标准曲线对比法的计算结果相比较。研究结果表明,标准曲线法的计算结果受人为主观影响误差较大;最小二乘法计算结果与实测数据拟合较好,但计算过程相对复杂;粒子群算法的求解精度最高,计算更快,具有良好的收敛性,是一种可靠的求解方法。