A numerical study of the effects of bottom and sidewall stiffeners on sloshing behavior considering roll resonant motion

In the present study, aiming to provide practical guidance for the structural design, the effects of the small bottom and sidewall stiffeners on sloshing loads in shallow and intermediate liquid conditions are investigated numerically. The interaction between the highly nonlinear free surface flow and a large number of small stiffeners is modeled based on the Moving Particle Semi-implicit method. 2D models of a rectangular clean tank and tanks with small sidewalls or bottom stiffeners are considered. For filling ratios ranging from 5% to 20%, harmonic angular motions with periods around the analytical sloshing resonant ones are applied. Simulation snapshots and computed pressure time series, pressure variation, wave run-up, maximum fluid velocity, and kinetic energy were provided to clarify the complex interaction between the small baffles and the shallow free surface flow. As a result, except for the tanks with bottom stiffeners in shallow liquid conditions, in which the response under actual resonant frequency should be considered, the response computed using the clean tank excited by its resonant period can be adopted as a conservative estimation for the sloshing loads of actual tanks with small stiffeners.     

复合地层中不同盾构掘进参数下碴土的颗粒特征研究

为定量研究盾构掘进复合地层时掘进参数对碴土颗粒级配特征的影响,以试验研究和统计分析为手段,得到掘进参数和颗粒特征之间的量化规律。首先,通过开展现场掘进试验,在不同有效推力和刀盘转速下实地采取碴样颗粒; 然后,对颗粒筛分结果进行统计分析,提出碴土颗粒级配预测方程; 最后,对级配预测方程的系数进行多元回归预测和影响性分析,定量研究方程系数和掘进参数之间的相关性。研究结果表明： 1）当盾构刀盘转速或盾构有效推力增大时,碴样特征粒径减小,碴土中的粗颗粒含量降低,细颗粒含量增大,碴样粒度分布范围增宽; 2）碴土颗粒的粒径存在上限,Rosin Rammler分布函数更加适用于盾构破岩颗粒级配特征分析; 3）在一定掘进参数范围内,盾构碴土颗粒级配方程系数的预测模型预测精度较高,能够较好地反映工程实践中掘进参数对方程系数的影响。     

基于试验数据的大跨度拱桥有限元模型修正

为了获取菜园坝长江大桥的基准有限元模型,结合Kriging代理模型和一种改进的粒子群优化算法,利用荷载试验数据对其初始有限元模型进行修正。首先,叙述模型修正和Kriging模型基本理论,在基本粒子群算法中引入交叉变异计算,提出一种改进的粒子群算法,并通过测试函数对改进的粒子群算法进行验证;其次,简要介绍菜园坝长江大桥荷载试验、荷载试验结果及初始有限元模型;最后,根据敏感性分析选定6个待修正参数,通过试验设计得到频率和位移关于修正的参数的样本,并建立有限元模型的Kriging代理模型以预测结构响应;以频率和位移的试验值和计算值残差为目标函数,分别利用基本粒子群算法和改进的粒子群算法在修正参数的设计空间内寻找目标函数的最小值,并对比分析模型修正的结果。结果表明：测试函数表明改进的粒子群算法具有较好的稳定性和成功率,并能获得更为精确的优化结果;建立的Kriging代理模型均方根误差较小,可以替代有限元模型预测结构频率和位移;经过模型修正,菜园坝长江大桥前5阶频率计算值与试验值相对误差均控制在5%之内;除个别测点外,位移相对误差均控制在10%以内;相比基本粒子群算法,改进的粒子群算法获得了更小的目标函数值,修正后的频率和位移的相对误差更小。     

融入偏好信息的列车运行过程多目标优化算法

列车运行过程优化是一个多目标、大滞后、非线性的极其复杂的优化问题.为了更好地解决上述问题,以列车能耗、舒适性、停靠准确性和运行时间为控制目标,以列车运动动力学方程为约束,建立了列车运行过程的多目标优化模型,提出了一种融入偏好信息的列车运行过程多目标遗传粒子群算法.提出的改进策略具有以下优点,在融入偏好信息的基础上通过控制粒子群中个体在解空间的分布能够更好地保持粒子群多样性,从而在进化过程中具有更明显的全局收敛的指向作用.仿真得到的速度距离曲线表明,在列车及其运行线路相同的情况下,本文所提出的算法性能较佳、寻优结果较好.     

纤维沥青混凝土的回弹模量试验研究

聚酯纤维沥青混凝土以其良好的路用性能得到人们广泛的认可,但纤维路面的设计参数尚不丰富。本文对5种纤维掺量沥青混凝土的圆柱体试件进行了静态无侧限单轴压缩试验研究,对抗压强度与回弹模量等设计参数的试验数据进行了分析,建立了回弹模量与纤维掺量的回归方程。试验结果表明,随着纤维掺量的增加,抗压强度与回弹模量均呈现出先增大后减小的趋势,并且分别在纤维掺量为0.2%和0.25%时达到最大值,由此可得出聚酯纤维沥青混合料的最佳纤维掺量应该在0.2%~0.25%之间。     

掺石灰对高液限土击实试验的影响

石灰作为一种有效的外掺剂可以显著降低其液塑限.为了指导施工,保证路基的压实质量,对不同灰剂量的石灰土进行标准击实试验,可得出最大干密度和最佳含水量随灰剂量的变化规律.     

自由设站放样测量在轨道交通连续刚构PC梁架设中的应用

芜湖轨道交通二号线是国内外首次采用跨坐式连续刚构PC梁的轨道交通项目,相较于国内外现有轨道交通项目的跨坐式简支PC梁,在梁体安装精度、线路平顺性控制、施工技术难度等方面均优于跨坐式简支PC梁。介绍了跨坐式连续刚构PC梁架设中自由设站放样测量控制网的测设原理、布点方法及布点形式,利用自由设站放样测量方法设站灵活、放样精度高等特点,结合自主研发的专用放样工装设备,解决了轻轨跨坐式连续刚构PC梁架设中测量精度不稳定、高程上桥困难、作业时遮挡严重、测量人员安全风险大、多榀梁连接平顺性差等测量难题,可将轨道梁中线偏差控制在3 mm以内,轨道梁中心变化率控制在1.5 mm/1.5 m以内,线路超高情况下的角度公差控制在±1/8°以内。实践证明:自由设站放样测量能够满足轨道PC梁精调施工定位的精度要求,且效率高、精度高、安全性高,在同类项目中具有推广和应用价值。     

Natural variability in exposure to fine particles and their trace elements during typical workdays in an urban area

Studies on the natural human exposures to fine particulate matter (PM_2.5) and their elements composition are practically non-existent in South America. In order to understand the natural exposure of the typical Brazilian population to PM_2.5 and their trace element composition, we measured PM_2.5 concentrations and collected mass on filters for nine continuous hours during a typical workday of volunteers. In addition, bus routes were performed at peak and non-peak periods, mimicking the routine activity of the population. Mean concentrations of PM_2.5 in the bus and car groups were similar while the fraction of BCe was higher for the bus group. For all routes, mean PM_2.5 concentrations were higher during peak than non-peak hours, with an average of 43.5 ± 33.1 μg m⁻³ and 14.3 ± 10.2 μg m⁻³, respectively. The trace elements S, K and Na originated mainly from vehicle emissions; Na was associated with the presence of biofuel in diesel. Toxic elements (Pb, Cr, Cu, Ni, Zn, Mn) were found at low levels as evident by the total hazard index that ranged from 2.15 × 10⁻⁰³ to 1.38 for volunteers. For all routes, the hazard index ranged from 2.25 × 10⁻⁰³ to 5.03. Average PM_2.5 respiratory deposition dose was estimated to be 0.60 μg/kg-hour for peak hours. Potential health damages to people during their movements and at workplaces close to the traffic were identified. Improvements in the design of the building to reduce the entrance of air pollutants as well as the use of filters in the buses could help to limit population exposure.     

基于改进LS-SVM算法的列车通信网络时延预测方法

由于通信网络诱导时延的存在会对列车牵引制动系统造成影响,因此对时延精准预测并实现补偿十分重要。提出了一种基于改进粒子群(PSO)算法优化的最小二乘法支持向量机(LS-SVM)算法对列车通信网络时延进行预测,搭建了列车网络控制系统半实物平台,使数据通过多功能车辆总线(MVB)进行传输,分别改变车辆控制单元(VCU)特征周期及负端口数量大小,以获取大量不同特性的时延数据。将数据分组后利用改进的PSO算法优化LS-SVM算法进行预测仿真。仿真结果表明,与传统的LS-SVM算法及Elman神经网络算法的预测方法相比,所提出的方法在列车通信网络的时延预测方面具有更好的快速性和准确性。     

基于旋转磁场磁化技术的铁道机车车辆转向架构架整体磁粉检测

以解决铁道机车车辆转向架构架整体磁粉检测为目的,介绍旋转磁场磁化的原理,对三维旋转磁场磁化方式用于磁粉探伤的优势和不足进行分析,为转向架构架磁粉检测设置专门的磁化装置以解决其磁粉检测的难题,并针对原理样机的功率消耗过大、磁悬液聚积问题给出了解决方案。