Inversion of potential field data

The problem of inversion of potential field data is a challenging one because of the difficulty in obtaining a unique solution. This paper identifies various types of nonuniqueness and argues that it is neither possible nor necessary to remove all categories of nonuniqueness. Some types of nonuniqueness are due to human limitations and choice and these would always persist. Listing all the solutions, imposing additional constraints on the acceptable solutions, a priori idealization, use of a priori or supplementary information, characterizing what is common to all the solutions, obtaining extremal solutions, seeking a distribution of all possible solutions, etc. are various responses in the face of nonuniqueness. It is shown that merely the form of nonuniqueness is changed by all these techniques. Some algorithms, which should be used to obtain the global minimum of the objective function are discussed. The conceptual commonality underlying seemingly different approaches and the possibility of nonunique interpretations of the same numerical results due to different axiomatic contexts are both elucidated.     

Bayesian inference for long-term prediction of significant wave height

This paper considers the problem of estimating long-term predictions of significant wave-height. A method which combines Bayesian methodology and extreme value techniques is adopted. Inferences are based on the Metropolis–Hastings algorithm implemented in an appropriate Markov Chain Monte Carlo scheme. The method is applied to obtain return values of extreme values of significant wave height collected on the northern North Sea. The results are compared with those obtained by Guedes Soares and Scotto [Guedes Soares, C. and Scotto, M.G., 2004. Application of the r-order statistics for long-term predictions of significant wave heights. Coastal Engineering, 51, 387–394].     

基于局部重力场建模的Tikhonov正则化点质量核径向基函数方法

针对点质量核径向基函数应用于局部重力场建模中的设计矩阵严重病态问题,本文引入Tikhonov正则化方法对传统点质量核径向基函数方程进行改造,建立了相应的正则化模型。通过模拟数据进行仿真实验,以传统格网化方法作为对比试验,利用"标靶法"确定两种模型的最优结构。实验结果表明:正则化点质量核径向基函数可以直接利用离散数据进行局部重力场建模。在两种模型的最优结构下,当实测数据无污染时,正则化方法达到与传统格网化方法相当的精度;当实测值中加入3 mGal的高斯白噪声时,正则化方法的精度获得了27.9%的提升。这说明本文方法可以应用于局部重力场建模中,且模型结构更优,抗干扰能力更强。     

On the use of Bayesian networks as a meta-modelling approach to analyse uncertainties in slope stability analysis

In this paper, we report on the use of Bayesian networks, BNs, learnt from data generated by physical and numerical models, to overcome to a certain degree a number of complications in traditional slope stability analyses that jointly consider the mechanical and hydraulic properties of soils. Discrete Bayesian networks resulted to be useful and efficient to acquire knowledge from simulated data and to identify significant factors by the combined use of backward inference and global sensitivity analysis. Further, BNs enable decision thresholds to be estimated quickly. Along with this, backward inference and global sensitivity analysis are performed in BNs at low computation costs. Moreover, under conditions in which knowledge is scarce, we show how a practitioner can be better informed using the proposed approach. All these previously under-reported modelling features in the specialised literature encourage the further application of the proposed approach to enhance slope stability analysis.     

地基沉降修正系数的Bayes概率推断

通过分析常规方法在沉降修正系数的选取中具有的定值性和随意性,引入建立在过去信息和现在样本信息之上的Bayes理论,结合某客运专线红黏土路基工程,提出用后验分布得到修正系数的取值范围。实例研究表明,用以往经验综合样本信息,估计修正系数的先验概率在某一区间上服从均匀分布。由现场载荷试验实测沉降量与理论计算沉降量分析所得的修正系数,将现场量测的沉降变形信息与先验信息结合起来,利用Bayes统计理论,由小样本试验数据推算得到修正系数的后验概率服从正态分布。对后验分布所得参数进行区间估计,得到该区域红黏土地基沉降修正系数的取值优化区间为 [1.0, 1.7],分析了不同荷载作用条件下沉降修正系数的概率分布模型。     

土地利用变化预测的案例推理方法

当前,基于案例的推理(Case-Based Reasoning,CBR)在解决复杂的地学问题时.对地学案例的表达和历史案例的相似性计算与推理存在明显缺陷,需要在CBR的表达模型和空间相似性计算与推理算法进行拓展.本文针对土地利用变化问题,首先在分析土地利用变化各种定量方法基础上,提出利用CBR进行土地利用变化分析的研究思路:其次,针对土地利用变化的空间特性及隐含的空间关系特性.给出土地利用变化案例的表达模型,案例间内蕴空间关系抽取算法,以及考虑案例间空间关系的CBR相似性推理模型:最后,进行珠江口区域土地利用变化的CBR方法试验.预测精度达到80%.为了进一步评价CBR方法对土地利用变化预测的有效性.在实例部分采用同样的实验数据进行贝叶斯网络的预测方法实验.由两种方法对比可知.CBR是从复杂到简单进行地学问题求解的一种有效方法.