Compactly Supported Correlation Functions

This article proposes compactly supported correlation functions, which parameterize the smoothness of the associated stationary and isotropic random field. The constructions are straightforward, and compact support is relevant for various ends: computationally efficient spatial prediction, fast and exact simulation, and appeal among practicioners.     

一种新钻孔孔位设计的最优化方法

本文给出了一种寻求新钻孔孔位的最优化方法。该方法将最优钻孔孔位问题转化为一个可计算的最优化问题,并用数论中的一致分布点列求解之。     

FEREBUS: Highly parallelized engine for kriging training

FFLUX is a novel force field based on quantum topological atoms, combining multipolar electrostatics with IQA intraatomic and interatomic energy terms. The program FEREBUS calculates the hyperparameters of models produced by the machine learning method kriging. Calculation of kriging hyperparameters ( θ and p ) requires the optimization of the concentrated log‐likelihood . FEREBUS uses Particle Swarm Optimization (PSO) and Differential Evolution (DE) algorithms to find the maximum of . PSO and DE are two heuristic algorithms that each use a set of particles or vectors to explore the space in which is defined, searching for the maximum. The log‐likelihood is a computationally expensive function, which needs to be calculated several times during each optimization iteration. The cost scales quickly with the problem dimension and speed becomes critical in model generation. We present the strategy used to parallelize FEREBUS, and the optimization of through PSO and DE. The code is parallelized in two ways. MPI parallelization distributes the particles or vectors among the different processes, whereas the OpenMP implementation takes care of the calculation of , which involves the calculation and inversion of a particular matrix, whose size increases quickly with the dimension of the problem. The run time shows a speed‐up of 61 times going from single core to 90 cores with a saving, in one case, of ～98% of the single core time. In fact, the parallelization scheme presented reduces computational time from 2871 s for a single core calculation, to 41 s for 90 cores calculation. © 2016 The Authors. Journal of Computational Chemistry Published by Wiley Periodicals, Inc.     

基于贝叶斯克里金的地下空间多源数据建模

利用贝叶斯克里金方法集成钻孔数据与地震解译数据估计地质层面的高程.将线性贝叶斯理论运用于克里金估计,把用于空间估计的数据分为2类,即把钻孔数据视为硬数据,把数据不确性较大的地震解译数据视为软数据,构建了一个考虑贝叶斯方法的地统计模型,用区域性变量理论研究这2类数据的空间变化特征.通过对某煤矿煤层表面高程的估计以及与普通克里金方法的估计值与误差的比较证明了贝叶斯克里金法能较好地考虑地质剖面和地震解译数据的不确定性,在地质层面高程的估计方面具有一定的优越性,是一种可行的考虑数据全局不确定性的多源数据集成建模方法.     

Local Gaussian Process Approximation for Large Computer Experiments

We provide a new approach to approximate emulation of large computer experiments. By focusing expressly on desirable properties of the predictive equations, we derive a family of local sequential design schemes that dynamically define the support of a Gaussian process predictor based on a local subset of the data. We further derive expressions for fast sequential updating of all needed quantities as the local designs are built up iteratively. Then we show how independent application of our local design strategy across the elements of a vast predictive grid facilitates a trivially parallel implementation. The end result is a global predictor able to take advantage of modern multicore architectures, providing a nonstationary modeling feature as a bonus. We demonstrate our method on two examples using designs with thousands of data points, and compare to the method of compactly supported covariances. Supplementary materials for this article are available online.     

Development of the Kriging Method with Application

This paper describes a modification of the kriging method for working with the square root transformation of a spatial random process. We have developed this method for the situation where the spatial process observed is not supposed to be stationary but the assumption is that its square root is a second order stationary spatial random process. Consequently this method is developed for estimating the integral of the process observed and finally some application of the method is given to data from an environmental radioactivity survey.     

Incorporation of local structure into kriging models for the prediction of atomistic properties in the water decamer

Machine learning algorithms have been demonstrated to predict atomistic properties approaching the accuracy of quantum chemical calculations at significantly less computational cost. Difficulties arise, however, when attempting to apply these techniques to large systems, or systems possessing excessive conformational freedom. In this article, the machine learning method kriging is applied to predict both the intra‐atomic and interatomic energies, as well as the electrostatic multipole moments, of the atoms of a water molecule at the center of a 10 water molecule (decamer) cluster. Unlike previous work, where the properties of small water clusters were predicted using a molecular local frame, and where training set inputs (features) were based on atomic index, a variety of feature definitions and coordinate frames are considered here to increase prediction accuracy. It is shown that, for a water molecule at the center of a decamer, no single method of defining features or coordinate schemes is optimal for every property. However, explicitly accounting for the structure of the first solvation shell in the definition of the features of the kriging training set, and centring the coordinate frame on the atom‐of‐interest will, in general, return better predictions than models that apply the standard methods of feature definition, or a molecular coordinate frame. © 2016 The Authors. Journal of Computational Chemistry Published by Wiley Periodicals, Inc.     

自适应Kriging近似模型及其在二维扩压器优化设计中的应用

将均匀设计方法、CFD技术、Kriging近似模型及小生境微种群遗传算法相结合发展了一种自适应全局优化设计方法.优化过程中综合考虑Kriging模型的预测值与预测标准差,引入了EI(Expected Improvement)函数得到校正点,解决了采用近似模型最优策略得到校正点带来的局部收敛问题.分别采用该方法和小生境微...     

基于Kriging模型和改进MCMC算法的随机有限元模型修正

针对待修正参数维数较高时,标准马尔可夫链蒙特卡罗MCMC (Markov Chain Monte Carlo)算法不易收敛、拒绝率高的问题,提出了基于Kriging模型和在MCMC中融合花朵授粉算法的修正方法.首先,以待修正参数作为输入,以应变模态作为输出,建立Kriging模型,通过蝙蝠算法确定Kriging模型的相关系数;然后,采用最大熵的贝叶斯方法估计参数的后验概率密度函数,将花朵授粉算法融入MH (M etropolis-Hasting)抽样算法,提高局部寻优和全局寻优能力;最后,通过三自由度弹簧-质量系统和三维桁架结构的数值算例验证所提模型修正方法,修正后参数相对误差均低于0.86％.结果 表明,所提方法修正后较高维参数的马尔可夫链能够快速收敛且样本接受率也有所提高,该方法也对随机噪声具有一定的鲁棒性.     

Enhancing water level prediction through model residual correction based on Chaos theory and Kriging

Hydrodynamic models based on the physical processes are indispensable tools for predicting water levels in ocean environment. Nonetheless, their accuracies are limited by various factors such as simplifying assumptions, complex ocean bathymetry, and so on. Residual correction, as one of the data assimilation techniques, can extract information from observation and assimilate it into a numerical model to correct the model output directly. Such correction is often performed in two steps: prediction of the model residuals at measured stations followed by spatial distribution at non‐measured locations. For long‐term residual forecast, the accuracy of prediction usually deteriorates with the forecast horizon. In addition to the residual correction at measurement locations, in this paper, we address the critical question as to how to effectively update outputs for computational points without measurements. We develop a hybrid data assimilation procedure, which combines a modified local model (MLM) and an approximated ordinary kriging (AOK). This technique improves the forecasts over a long horizon over the entire computational domain. Using the proposed residual correction technique, the hybrid procedure is examined on a case study of Singapore Regional Model for correcting the water level outputs at locations with and without measurements. In order to provide a comparison, the analysis is carried out throughout prediction horizon of model residuals, tidal residuals, and sea level anomaly, respectively. The comparisons show that the proposed method can successfully assimilate and forecast all variables. Results indicate that resulting prediction accuracy can be significantly improved for all locations of interest independently of the forecast horizon. Copyright © 2014 John Wiley & Sons, Ltd.