基于多元自回归样条的液化侧移灾害评估模型

为了将液化侧移计算方法本土化,本文基于Youd等人收集的液化侧移数据库,采用擅长处理复杂参数维度问题的多元自适应样条回归法（MARS）建立液化侧移计算模型。经验证,该方法建立的模型精度和可靠性都可以通过检验。本文主要内容包括：（1）在本土化过程中,本文选取峰值加速度PGA作为地震动参数,替代现有模型中的M和R,计算结果并未影响液化侧移计算精度;（2）在MARS方法进行建模过程中,对临空和缓坡两种情况的影响因素进行重要性程度计算和筛选;（3）建立两种情况下的液化侧移计算模型,并用F检验法和t检验以及新西兰2010—2011年系列地震中液化侧移数据进行检验,证明本文方法的合理性和可靠性。本文的结果能满足我国相关规范、工程师以及岩土地震工程防灾减灾等方面的需求,还可为地震液化侧移灾害小区划提供理论基础。     

钢筋-沥青隔震层非线性震动特性研究

为了研究钢筋-沥青隔震层的材料与几何非线性震动特性,对钢筋-沥青隔震结构进行了振动台试验,分析了结构在不同加速度幅值输入时的隔震性能。同时,在钢材的应力-应变关系基础上,考虑构件竖向力的二次矩效应,提出了钢筋-沥青隔震层骨架曲线的计算方法,并结合双线型恢复力模型的滞回规则,建立了相应的恢复力模型。利用建立的恢复力模型,编制程序进行了弹塑性时程分析,并与试验结果作对比。结果表明:当钢筋-沥青隔震层的受力钢筋进入弹塑性阶段时,隔震层能够吸收大部分地震能量,显著降低上部结构的地震反应;弹塑性时程分析结果与试验结果吻合较好,由此建立的恢复力模型准确并且适用,可为钢筋-沥青隔震层的工程设计与非线性地震反应分析提供参考。     

Active channel width as a proxy of sediment supply from mining sites in New Caledonia

Although the channel morphology of upland fluvial systems is known to be strongly controlled by sediment supply from hillslopes, it is still difficult to isolate this effect from the other controlling factors of channel forms, such as the sediment transport capacity (depending notably on the size of the catchment) and local conditions (e.g. confinement, riparian vegetation, valley-floor slope). The rivers in New Caledonia offer an interesting field laboratory to isolate the morphological effect of contrasted sediment supply conditions. Some of these rivers are known to be highly impacted by the coarse sediment waves induced by the mining of nickel deposits that started in the early 1870s, which was particularly intensive between the 1940s and 1970s. The propagation of the sediment pulses from the mining sites can be traced by the presence of wide and aggraded active channels along the stream network of nickel-rich peridotite massifs. A first set of 63 undisturbed catchments in peridotite massifs distributed across the Grande Terre was used to fit a classic scaling law between active channel width and drainage area. A second set of 86 impacted sites, where the presence of sediment waves was clearly attested by recent aerial imagery, showed systematically wider active channels, with a width ratio around 5 (established from the intercept ratio of width–area power laws). More importantly, this second set of disturbed sites confirmed that the residual of active channel widths, computed from the scaling law of undisturbed sites, is statistically positively related to the catchment-scale relative area of major mining sediment sources. It is therefore confirmed that the characterization of sediment supply conditions is crucial for the understanding of spatial patterns of active channel width, and this should be more thoroughly considered in morphological studies of rivers draining environments with contrasted geomorphic activities on hillslopes. © 2018 John Wiley & Sons, Ltd.     

Test and analysis of reinforced concrete (RC) precast shear wall assembled using steel shear key (SSK)

Reinforced concrete (RC) precast shear walls are extensively applied in practical engineering, owing to their fast construction speed. However, because of the transport conditions, RC precast shear walls have to be separated into small wall segments during the factory prefabrication procedure before being assembled on site. Typically, wet-type jointing methods are adopted to link the segments, which is time-consuming and results in unreliable post-pouring area strength. To overcome this problem, the novel scheme of the steel shear key (SSK) featuring steel shear panels and combined fillet and plug welding is proposed. Three RC precast shear wall specimens with different linking strength, termed as weakened SSK wall, standard SSK wall, and strengthened SSK wall, respectively, and an integrated shear wall specimen were designed. Quasi-static cyclic loading was applied to investigate the specimens' dynamic properties. The test results suggest the prefabricated wall segments equipped with SSKs showed reliable stiffness and bearing capacity and were improved in energy dissipation ability, compared with conventional shear walls. As the shear stiffness and number of equipped SSKs increased, the specimens exhibited higher strength, but their ductility and energy dissipation were slightly decreased. Most importantly, the standard SSK wall specimen could achieve satisfactory bearing capacity and deformability and is thus recommended for precast building structures. Finite element method (FEM) models were established to validate the test results, and parametric study analysis was conducted based on the coupling ratio of the SSK walls. Finally, an appropriate coupling ratio range is recommended for practical engineering applications.     

双向地震作用下锯末混合土场地动力特性分析

在北京工业大学振动台台阵系统上开展了一系列锯末混合土地基自由场振动台模型试验,试验中模型箱采用装配式连续体刚性模型箱,试验中输入地震动时程采用El Centro地震动记录、Taft地震动记录和天津地震动记录,地震动输入方向分为水平单向和水平双向。文中,重点考察了双向地震动输入下锯末混合土模型场地的动力特性及其变化规律,主要指标包括模型场地地震动反应的峰值加速度及其动力放大系数、加速度时程及其傅氏谱。试验结果表明:随着输入地震动强度的增大,同一测点反应的峰值加速度总体上在增大,而其加速度动力放大系数总体上呈现减小的趋势,反应的频谱组成从较高频率向较低频率移动;双向地震作用下锯末混合土模型场地的动力变化规律与单向地震作用下较为一致。     

地磁低点时间与经度的相关性分析

地磁低点时间在空间分布上的主要特征是随着经度变化而变化。为了求出这种变化关系的解析式,首先使用国家地磁台网的数据计算甘肃省近三年13个地磁台站的低点时间序列和低点时间的期望,并绘制了低点时间序列的频次分布图和概率密度曲线;然后对低点时间与台站经度的关系做线性回归,并指出这个线性回归方程应无限趋近于“格林尼治时间-经度”关系式。文章第一次提出台站经度与地磁Z分量低点时间的关系式,并提出它的趋近方程。各台站低点时间的期望就是地磁低点时间的正常背景值,它是判定地磁低点时间异常与否的参考标准。本文的工作完善和发展了丁鉴海先生的地磁低点位移法。     

基于车桥耦合振动的常用简支空心板梁桥冲击系数分析

通过Universal Mechanism和ANSYS软件相结合建立了车桥耦合振动模型,考虑车型、行车速度、车重和桥梁频率等因素,对工程常用的4种标准跨径装配式混凝土空心板梁桥进行了车桥耦合振动模拟。研究表明:冲击系数随车速增加而上升;随车重增加而增大;随桥梁频率的增加而降低。在多种因素中,车速对冲击系数影响最为明显。     

基于振动台试验的移动通信节点机房蓄电池组地震易损性分析

关键移动通信设备的地震易损性是城市移动通信系统地震灾害预测的基础。本文以移动通信系统节点机房中常用的蓄电池组作为研究对象,通过振动台试验来研究蓄电池组的抗震性能,并获得其地震易损性。试验过程中,首先采用FEMA P695建议的方法从美国PEER强震数据库中遴选出对蓄电池组较为危险的地震动记录作为试验激励,随后逐步增大各条地震动记录的幅值,基于振动台进行增量动力分析。试验后根据蓄电池组的地震损伤特征,定义了其损伤指标和损伤水平,并基于振动台试验数据分析获得了典型蓄电池组的地震易损性曲线。     

机械连接预应力实心方桩承台节点抗震性能试验研究

预应力管桩在高烈度抗震设防地区已禁用,而普通钢筋混凝土方桩成本较高。首先针对卡扣式机械连接预应力混凝土实心方桩、PHC管桩和普通钢筋混凝土方桩与承台处的节点进行拟静力试验,并利用试验数据针对Park-Ang损伤模型进行了修正,进行了损伤指数分析。研究结果表明:预应力方桩比普通钢筋混凝土方桩的抗裂性能明显好,而预应力混凝土实心方桩承台节点耗能能力和延性低于预应力管桩和普通钢筋混凝土方桩;截桩施工导致桩端部损失部分预应力,相比较未截桩试件屈服荷载和极限荷载均减小,节点裂缝开展较早,损伤较未截桩试件严重;增配普通钢筋试件的滞回曲线比较饱满,水平承载能力和位移延性得到提升,节点抗震性能得到提升与改善;预应力混凝土实心方桩损伤前期较小,而一旦发生损伤后,损伤指数发展较快。     

缩短核磁共振测定冻土未冻水含量实验时间的方法

为利用核磁共振(NMR)技术更快地测试得到冻土中未冻水的含量, 首先对传统的单试样测试方法进行了分析, 探究了其实验时间过长的原因. 然后, 基于传热学中的集总参数法, 研究了小体积试样在空气中以自然对流传热为冷却方式的温度变化特征. 根据试样温度与时间的对应关系, 提出了多试样连续测试法. 最后, 以一种黏土为例, 分别利用多试样连续测试法和单试样测试法测量了土样冻结过程中NMR信号强度随温度变化的关系. 实验结果表明: 二者的测量结果非常接近, 但连续测试法耗费的时间非常少. 综合比较两种利用NMR技术所耗费的时间和实验结果的精确程度, 多试样连续测试法是可行的.