On the applicability of neural networks for soil dynamic amplification analysis

Artificial Neural Networks (ANN) have gained a solid status as a tool for modeling complex phenomena in different areas of research and engineering practice. In this paper, their applicability to estimate the mapping of seismic acceleration from bedrock to free surface in a complex soil profile is explored. Such a use is intended to serve as a hypothesis-free alternative to the dynamic amplification analysis, which is currently based on geophysical and soil dynamics procedures. Were the neural networks to be useful to such a mapping, they could in principle be employed for several purposes, such as soil identification using instrumental data, design of early warning systems and estimation of probabilistic spectra via Monte Carlo simulation, in which the ANN act as an efficient solver surrogate. The conditions under which these ambitious purposes can be reached are discussed. Two classes of multi-layer perceptrons were tested, which are characterized by time-independent and time-dependent connections. It is shown that the first class of networks is useful for response spectrum mapping, while the second performs very well in the assessment of free-surface time series. It arises as the main conclusion that the most promising perspective of application of ANN in this respect is for the estimation of probabilistic free-surface spectra, which is an important goal for the modern trend of reliability-based aseismic design. The limitations to the other said applications are also highlighted.     

不规则三角网(TIN)在流域坡面汇流分析中的应用

数字高积模型（DEM）的地形解析能力在提取地貌特征中的应用已经十分广泛，流域的坡面汇流分析及径流的数字模拟工作也随之展开。但是目前分析中常用的方格网络模型、等高线网络模型和不规则三角网络模型在应用中都存在着很大的缺陷，因此本文提出了一种融合方格网络和等高线网络的DEM－TIN网络模型，该模型具有上述三种模型的优点。同时以DEM数据和数字线划图（DLG）为基础，设计了这种模型的简易生成算法。     

神经网络串行训练并行处理

作者在文中提出一种神经网络串行训练处理方法，它能提高神经网络学习收敛的速度，并达到提高训练精度的要求，此外，该算法还可依据实际问题的需求，自适应地扩展神经网络的容量，使得可最大限度地充分利用已知信息，达到增强神经网络预测结果可靠性的能力，作者在文中还给出了本文算法运用于大庆油田甘开发区块储层油气预测应用实例，效果良好。     

基于GIS技术的秦淮河流域水系分维研究

流域水系是一种自然分形,基于GIS技术在MapInfo和MapBasic平台下,采用编程手段,通过改变粗视化程度求维数的数盒子法求取了秦淮河流域水系的盒维数,并在此基础上讨论了水系分维的意义。指出水系的分维不仅反映了该流域水系的发育程度,而且和流域的地貌系统、土地利用以及道路网都有着密切的关系。     

Hydrological mapping from SPOT imagery

Four SPOT images, one panchromatic and three multispectral, were studied to evaluate their usefulness for hydrological mapping in an arid environment. Simple visual methods of interpretation were used to plot the drainage network independently from each test image. The results were compared with the drainage network shown on the topographic map of the area. Results show that with the panchromatic image, over 90 per cent of the map content could be discerned, identified, and plotted. The accuracy of plotting the drainage system from the multispectral images seems to depend on the season in which data had been acquired. However, in all cases, the success rate is not less than 70 per cent. Comparison with Landsat Thematic Mapper (TM) imagery showed that SPOT system is superior in hydrological mapping.     

Focus-of-attention techniques in the automatic interpretation of seismograms

The focus-of-attention techniques implemented in SNA2, a knowledge-based system for seismogram interpretation, are presented. They consist of data compression of the input digital records, scanning of the compressed traces to detect candidate seismograms and extraction of seismogram features. A criterion is given to rate the clarity of seismograms; the clarity defines the order in which the system will consider them to build up the interpretation. The proposed techniques are simple and fast; they allow quick rejection of noise and focussing the attention of the system on the portions of traces containing relevant information.     

The fractal geometry of tidal-channel systems in the Dutch Wadden Sea

Horton's hierarchical and fractal analysis of channel circumference reveals that tidal-channel systems in the Dutch Wadden Sea have similar branching patterns. Channel systems have the same characteristics as three- to four-times branching networks. The branch lengths of these channels decrease logarithmically. The channel systems can be regarded as statistical self-similar fractal networks, considering the natural variability in branch lengths and channel positions. Branching of channels does not continue below the 500 m scale. The channel-system circumference length is logarithmically related to the tidal prism and drainage area. The similarity of the channel systems, regardless of their size, relative amount of intertidal flats, and tidal amplitude, points to a self-organising nature. All processes depend on the feedback between morphology and hydrodynamics. At first sight, the channel systems can be regarded as an ebb-driven drainage network, governed by erosion. However, flood-dominated net sedimentation occurs in large parts of the drainage basins and modifies the ebb-driven network. The complex interaction of hydrodynamic and morphodynamic processes in tidal basins limits the applicability of process-based models. Behaviour-oriented modelling has a wide applicability and can be improved using the fractal geometry as the dynamical equilibrium morphology. The fractal-network geometry can also be used for stochastic reconstructions of fossil tidal-channel systems, when only limited observations are available.     

深度融合网结合条件随机场的遥感图像语义分割

为了充分利用遥感图像中丰富的细节信息和上下文信息,提高图像语义分割精度,提出一种深度融合网结合条件随机场模型的遥感图像语义分割方法。方法在全卷积神经网络框架中增加反卷积融合结构,搭建深度融合DFN (Deep Fusion Networks)网络,通过深层网络自动获取多尺度特征,避免人工设计和选择特征,提高模型的泛化能力;同时借助反卷积融合结构,利用多尺度信息,将浅层细节信息和深层语义信息相融合,提高模型的处理精度。由全连接条件随机场引入空间上下文信息,更好地定位边界,得到最终的语义分割结果。在遥感图像数据集上的实验结果显示:(1)随着不同尺度细节信息的融入,结果的边缘轮廓越精确、接近标签图像;(2)增加了空间上下文信息后,语义分割结果边缘更细化、准确,精度更高。实验表明,该方法可以有效提高遥感图像语义分割的精度,改善结果的过平滑现象。     

神经网络法在寻找“隐伏矿（盲矿）”中的应用机理探讨

随着计算机应用技术的不断提高，神经网络作为一个数据处理结构应运而生。传统的遥感找矿方法是通过日视解译圈定“靶区”，但在“隐伏矿(育矿)找矿”项目中却遇到了难题，很难找到相关弱信息，利用神经网络的模拟人脑功能来提取矿化相关弱信息寻找“隐伏矿(育矿)”是可能的。在此仅仅对其应用做简单的介绍，希望对今后的遥感找矿产生一定的提示作用。     

最短路径射线追踪方法及其改进

综述了用网络最短路径算法求解地震射线追踪问题的原理、方法技术以及存在问题和改进措施。特别介绍了作者在最短路径算法基础上，提出的动态网络最短路径地震射线追踪方法。该方法先采集从炮点到整个模型所有节点上的初至旅行时，其中，在一个单元内，对相邻每对已计算出最小旅行时的节点进行线性插值，并利用Fermat原理计算未知节点的最小旅行时；然后，利用同样的方法，从接收点开始，反向追踪炮点到接收点的射线路径、该方法能适于各种复杂的非均匀介质，极大地提高了射线追踪的精度。