三维大地电磁自适应正则化有限内存拟牛顿反演

有别于传统基于梯度信息的反演方法在正则化约束中用总梯度逼近海塞逆矩阵的技术,本文将正则化约束问题的数据拟合项和模型光滑项分开考虑,只利用数据拟合函数的梯度信息对数据拟合项的海塞矩阵进行逼近,通过求解类高斯牛顿下降方向方程得到不依赖前几次迭代正则化因子的更精确下降方向,在求解当前迭代下降方向的过程中,通过保证右端项中两个向量的二范数在同一数量级的原则,实现了正则化因子的自动更新.对理论模型的试算表明这种自适应正则化反演方案可以在拟牛顿反演框架下基本达到OCCAM的算法稳定性,反演结果对初始模型依赖性较小,同时又无需在一次迭代中多次搜索最佳正则化因子.本文还基于此算法讨论了大地电磁各参数对于反演结果的影响,由于本文的反演结果能得到充分的正则化约束,因而在此框架下讨论阻抗和倾子在反演中的作用相对更为客观.

     

REJECTING VIOLENCE ON THE LANDSCAPE IN LAWRENCE,KANSAS*

ABSTRACT. Lawrence, Kansas, survived a tumultuous beginning. The young town endured attacks twice, in 1856 and 1863. The second raid, by the guerrilla William Quantrill's troop of more than 400 men, resulted in the deaths of 143 citizens. Lawrence serves as an example of how Americans memorialize unconventional warfare, targeted at citizens, in a material and permanent fashion on the landscape. Small and obscurely placed memorials fill the town, to the point that they have become ordinary. The memorialized landscapes of these tragedies thus display ambivalence toward the past and symbolically reject the loss of lives despite this era's high position in the literature and archival history of the town. Additionally, Lawrence has found alternative sources for its historical identity that do not reflect these tragedies but instead celebrate the city's pioneer establishment. In this article I use a set of methods for reading the memorialized landscape that includes archival and landscape analysis and uncovers the processes that have led to this town's understated and ambivalent memorialization and identity.     

一种工业计算机局部网络通信协议

介绍了一种用于工业现场的计算机局部网络通信协议．详细介绍了网络的硬件结构和与之相配合的软件设计．同时，对其中关键技术如介质存取控制方式，数据的ＤＭＡ存取，缓冲区的分配与管理等作了重点说明．     

Structural vibration control by shape memory alloy damper

A damper device based on shape memory alloy (SMA) wires is developed for structural control implementation. The design procedures of the SMA damper are presented. As a case study, eight such SMA dampers are installed in a frame structure to verify the effectiveness of the damper devices. Experimental results show that vibration decay of the SMA damper controlled frame is much faster than that of the uncontrolled frame. The finite‐element method is adopted to conduct the free and forced vibration analysis of the controlled and uncontrolled frame. The experimental and numerical results illustrate that the developed SMA dampers are very effective in reducing structural response and have great potential for use as efficient energy dissipation devices with the advantages of good control of force and no lifetime limits, etc. Copyright © 2003 John Wiley & Sons, Ltd.     

现场多功能定点观测系统

现场多功能定点观测系统是为潮汐河口及沿海地区进行水文测验而研制的。它由传感器组件、压力仓、电子电路、数据采集和处理装置等组成。传感器组件有流速、流向、温度、压力、电导率和含沙量等传感器，能测量流速等６种水文参数。该系统的使用既可长期定点锚泊测量，也可短期随船测量。长期锚泊测量能适应恶劣环境。     

思念裴文中院士

裴文中教授是我国地质学界前辈中最富传奇色彩的杰出科学家之一 ,因发现周口店“北京人”而享誉科学界。他的聪明才智和对科学的执着追求 ,终于把伟大的科学发现带给了科学界和广大民众。他的业绩和关心贫苦的精神深深影响了他的同事和学生也影响了同时代的中国地质学家。为纪念这位伟大的第四纪地质学家、古脊椎动物学家和史前考古学家逝世 2 0周年 ,作者怀着无限崇敬的心情 ,通过日常接触和交往中的一些小故事 ,介绍了裴文中教授的人格魅力 ,他留下的这些精神财富无疑是对我们为社会和科学事业做出更多贡献的巨大推动力。在这篇纪念短文中 ,作者希望大家从裴文中教授之所以在其科学生涯中取得成就 ,和在中国地质学界受敬仰的点点滴滴中得到启示。     

Patterns of disaster commemoration in long‐term recovery

The postdisaster landscape is replete with memorials that help communities collectively remember destructive events and recover psychologically. Although commemoration is intrinsic to all stages of recovery, little research from the disaster‐science field engages memorial texts across disasters. Meanwhile, a rich body of work on memorials and their functions exists in the cultural geographic tradition. Drawing from this literature, the current study examines a sample of U.S.‐based memorials to discern patterns within the postdisaster commemorative landscape. This research leverages discourse analysis to interrogate the meanings and mechanics of postdisaster memory work. Findings revealing that disasters catalyze remembrances that remake places, postdisaster memorial texts construct wide‐ranging degrees of intimacy, and memorials distilling survivor memories impel community recovery differently than memorials that reconstruct imagined pasts. These identified patterns in postdisaster commemoration enable further systematic exploration of memory work in the long‐term recovery process.     

Application of deep learning technique to the sea surface height prediction in the South China Sea

A deep-learning-based method, called ConvLSTMP3, is developed to predict the sea surface heights(SSHs).ConvLSTMP3 is data-driven by treating the SSH prediction problem as the one of extracting the spatial-temporal features of SSHs, in which the spatial features are "learned" by convolutional operations while the temporal features are tracked by long short term memory(LSTM). Trained by a reanalysis dataset of the South China Sea(SCS), ConvLSTMP3 is applied to the SSH prediction in a region of the SCS east off Vietnam coast featured with eddied and offshore currents in summer. Experimental results show that ConvLSTMP3 achieves a good prediction skill with a mean RMSE of 0.057 m and accuracy of 93.4% averaged over a 15-d prediction period. In particular,ConvLSTMP3 shows a better performance in predicting the temporal evolution of mesoscale eddies in the region than a full-dynamics ocean model. Given the much less computation in the prediction required by ConvLSTMP3,our study suggests that the deep learning technique is very useful and effective in the SSH prediction, and could be an alternative way in the operational prediction for ocean environments in the future.     

Shaking table tests of steel frame with superelastic Cu–Al–Mn SMA tension braces

Shaking table tests are performed on a one‐bay one‐story steel frame with superelastic Cu–Al–Mn shape memory alloy (SMA) tension braces. The frame is subjected to a series of scaled ground motions recorded during the 1995 Kobe earthquake, Japan. The test results demonstrate that the SMA braces are effective to prevent residual deformations and pinching. It is also shown that the time history responses observed from the shaking table tests agree well with the numerical predictions using a rate‐independent piecewise‐linear constitutive model calibrated to the quasi‐static component tests of the SMA braces. This suggests that the loading rate dependence of Cu–Al–Mn SMAs as well as the modeling error due to the piecewise linear approximation can be neglected in capturing the global response of the steel frame. Numerical simulations under a suite of near‐fault ground motion records are further performed using the calibrated analytical models to demonstrate the effectiveness of the SMA braces when the variability of near‐fault ground motions is taken into account. A stopper, or a deformation restraining device, is also proposed to prevent premature fracture of SMA bars in unexpectedly large ground motions while keeping the self‐centering capability in moderate to large ground motions. The effectiveness of the stopper is also demonstrated in the quasi‐static component and shaking table tests. Copyright © 2015 John Wiley & Sons, Ltd.     

工程结构的形状记忆合金超弹性阻尼减振技术研究进展

介绍了形状记忆合金(SMA)超弹性阻尼减振的机理，综述了国内外形状记忆合金(SMA)减振装置用于工程结构振动控制的最新进展。大量的研究结果表明形状记忆合金阻尼减振技术在土木工程结构振动控制方面具有比较好的应用前景。