Site characterization model using least‐square support vector machine and relevance vector machine based on corrected SPT data (Nc)

Statistical learning algorithms provide a viable framework for geotechnical engineering modeling. This paper describes two statistical learning algorithms applied for site characterization modeling based on standard penetration test (SPT) data. More than 2700 field SPT values (N) have been collected from 766 boreholes spread over an area of 220 sqkm area in Bangalore. To get N corrected value (N_c), N values have been corrected (N_c) for different parameters such as overburden stress, size of borehole, type of sampler, length of connecting rod, etc. In three‐dimensional site characterization model, the function N_c=N_c (X, Y, Z), where X, Y and Z are the coordinates of a point corresponding to N_c value, is to be approximated in which N_c value at any half‐space point in Bangalore can be determined. The first algorithm uses least‐square support vector machine (LSSVM), which is related to a ridge regression type of support vector machine. The second algorithm uses relevance vector machine (RVM), which combines the strengths of kernel‐based methods and Bayesian theory to establish the relationships between a set of input vectors and a desired output. The paper also presents the comparative study between the developed LSSVM and RVM model for site characterization. Copyright © 2009 John Wiley & Sons, Ltd.     

Saldanha Bay,South Africa I: the use of ocean colour remote sensing to assess phytoplankton biomass

The efficacy of ocean colour remote sensing in assessing the variability of phytoplankton biomass within Saldanha Bay is examined. Satellite estimates of chlorophyll a (Chl a) were obtained using the maximum peak-height (MPH) algorithm on full-resolution (300?m) data from the Medium Resolution Imaging Spectrometer (MERIS). Subsurface Chl a maxima often occur within Saldanha Bay below the mean detection depth of the satellite (1.5?m) during periods of thermal stratification. Consequently, the MPH product was poorly correlated to in situ data from 4?m depth (r² and average relative percentage difference [RPD] of 0.094 and 53% respectively); however, the coefficient of determination was much improved if limited to in situ data collected under conditions of mixing (r² and RPD of 0.869 and 89%, respectively). Composites of monthly MPH Chl a data reveal mean concentrations consistent with in situ seasonal trends of phytoplankton biomass, confirming the capability of the MPH algorithm to qualitatively resolve surface Chl a distribution within the bay.     

ANN在海洋预报中的应用

人工神经网络通过神经元之间的相互作用来完成整个网络的信息处理,具有自学习和自适应等一系列优点,因而用它来进行海洋预报是可行的。针对海洋预报问题,初步建立了基于神经网络的预报分析系统,给出了应用实例。     

福建及其周边地区地震活动人工神经网络模型的构建

本文通过对福建及其周边地区地震活动人工神经网络模型的构建,研究了人工神经网络方法在基于该区域地震活动性指标的地震分析预报中的应用。选用含一个中章层的前向神经网络模型,并采用与之相适应的ＢＰ算法,以该地区１９７１～１９９７年的地震活动性资料为基础,用神经网络进行实际计算、分析和检验。结果表明：神经网络模型对福建及其周边地区地震震级的预测检验效果较好的,可以在一定精度范围内使震级预测的内符率达１００％     

浅述多元数据处理的一种简化拟合形式及算法剖析

在实际工程试验与研究中 ,多元数据处理经常遇到 ,根据 Weierstrass定理 ,采用多项式拟合是一种常见的数据处理方法。对于多元多项式的系数求解问题 ,可以归结为超定方程组的求解问题。该文以超定方程组的 SVD求解算法为核心 ,阐述了在工程实际中关于多元数据处理的一种应用方法 ,并尝试给出多元多项式拟合的一种简化方法     

三种遥感反演海岸线方法的实验对比分析

在分析海岸线提取算法基础上,以环渤海的Landsat7 ETM+影像为数据源,以基岩海岸线和淤泥质海岸线两种典型的海岸线为研究对象,采用Sobel边缘检测算法、阈值分割法和K最邻近分类算法进行海岸线提取实验。通过实验对比分析,归纳和总结了针对不同类型海岸线提取算法的适用性。     

确定连续系统参数的时域方法

提出一种确定连续系统参数的时域方法。该方法利用方波输入和系统稳定性，离析出输出采样信号与结构参数的内在联系，推导出确定连续系统结构参数的算法．指出该算法与辨识方法相结合，可实现对含随机干扰的连续系统结构参数的估计．     

海上固定平台雷达测波技术及典型数据处理算法

首先总结了国内外海上固定平台使用的多种非接触式波浪测量方法,比较了不同测波方法的适用性。针对不同的测量方法给出了常用的传感器与设备,同时提供了每种设备的主要生产销售商。然后,对单点测波雷达和X波段雷达测波算法进行了综述,给出了波面的时间序列和雷达影像的时空序列两种数据的处理过程。最后,提出了波浪测量技术存在的一些问题,以期为我国海洋环境监测与观测系统的完善提供参考。     

Multiple-Step Predictive Control for Offshore Structures

—Ocean wave propagation is slow,visible and measurable,so a wave theory can be used to approxi-mately predict the imminnent wave force on an offshore structure based on measured,real-time wave elevation nearthe structure.This predictability suggests the development of a more efficient algorithm,than those that have beendeveloped for structures under wind and seismic loads,for the active vibration control of offshore structures.Thepresent study delveops a mutiple-step predictive optimal control(MPOC)algorithm that accounts for multiple-step external loading in the determibation of optimal control forces.The control efficiency of the newly developedMPOC algorithm has been investigated under both regular(single-frequency)and irregular(multiple-frequency)wave loads,and compared with that of two other well-known optimal control algorithms:classical linear optimalcontrol(CLOC)and instantaneou optimal control(IOC).     

基于阈值区间的海洋锋面提取模型

A model （Bayesian oceanic front detection, BOFD） of sea surface temperature （SST） front detection in satel- lite-derived SST images based on a threshold interval is presented, to be used in different applications such as climatic and environmental studies or fisheries. The model first computes the SST gradient by using a Sobel algorithm template. On the basis of the gradient value, the threshold interval is determined by a gradi- ent cumulative histogram. According to this threshold interval, front candidates can be acquired and prior probability and likelihood can be calculated. Whether or not the candidates are front points can be deter- mined by using the Bayesian decision theory. The model is evaluated on the Advanced Very High-Resolution Radiometer images of part of the Kuroshio front region. Results are compared with those obtained by using several SST front detection methods proposed in the literature. This comparison shows that the BOFD not only suppresses noise and small-scale fronts, but also retains continuous fronts.