Applying Bayesian belief networks to health risk assessment

The health risk of noncarcinogenic substances is usually represented by the hazard quotient (HQ) or target organ-specific hazard index (TOSHI). However, three problems arise from these indicators. Firstly, the HQ overestimates the health risk of noncarcinogenic substances for non-critical organs. Secondly, the TOSHI makes inappropriately the additive assumption for multiple hazardous substances affecting the same organ. Thirdly, uncertainty of the TOSHI undermines the accuracy of risk characterization. To address these issues, this article proposes the use of Bayesian belief networks (BBN) for health risk assessment (HRA) and the procedure involved is developed using the example of road constructions. According to epidemiological studies and using actual hospital attendance records, the BBN-HRA can specifically identify the probabilistic relationship between an air pollutant and each of its induced disease, which can overcome the overestimation of the HQ for non-critical organs. A fusion technique of conditional probabilities in the BBN-HRA is devised to avoid the unrealistic additive assumption. The use of the BBN-HRA is easy even for those without HRA knowledge. The input of pollution concentrations into the model will bring more concrete information on the morbidity and mortality rates of all the related diseases rather than a single score, which can reduce the uncertainty of the TOSHI.     

基于贝叶斯方法的深层充电效应风险模型评估

高能电子穿透航天器并在其内部沉积电荷从而引发深层充电效应,是导致卫星故障的重要因素之一.为了评估深层充电效应诱发卫星异常的风险,本文基于贝叶斯方法,使用一颗地球同步轨道卫星的异常数据和GOES-8卫星的电子通量探测数据,计算了不同能量阈值及累积时间的电子注量、不同卫星配置下模拟仿真的沉积电荷,并分别与卫星异常建立一系列概率风险模型.本文从模型中随机抽样得到模拟异常,并与实测异常构造混淆矩阵以评估模型拟合优度,结果表明>1.0 MeV电子3日累积注量-卫星异常概率风险模型为该卫星最优模型.本文利用最优模型对该卫星深层充电效应风险进行了计算,在>1.0 MeV电子3日累积注量达到2.0×10¹⁰cm^-2·sr^-1时,该卫星发生深层充电异常的平均后验概率为27%,且95%最小可信值为22%.根据最优模型,我们对该卫星最可能导致异常的部件的材料和结构等特征做出了推断.

     

基于优化聚类的人工源电磁法数据信噪分离方法

为了降低强电磁干扰对人工源电磁法(Controlled Source Electromagnetic Method, CSEM)有效信号的影响, 改善CSEM实测数据处理结果因人而异且效率低的不足, 本文针对CSEM有效信号周期性特征提出了一种加权自适应带宽均值漂移聚类(Weighted Adaptive Bandwidth Mean-Shift Clustering, WAB-MSC)信噪分离方法.首先在传统均值漂移聚类(Mean-Shift Clustering, MSC)算法的基础上增加核函数, 降低处理结果对带宽选择的敏感度, 提高算法的稳健性; 其次结合实测CSEM数据的分布特征提出了一种基于局部密度梯度的带宽估计方法, 实现了自适应带宽选择; 最后通过仿真数据与实测数据对本文方法进行了验证, 结果表明: 本文方法能有效消除强电磁干扰对CSEM数据的影响, 最大程度保留受噪声影响较小或未受噪声影响的数据, 提高数据信噪比, 降低强干扰噪声对CSEM初始资料的影响程度, 获得更为真实的地电响应模型, 为后续数据处理提供保障.

     

Nonparametric catchment clustering using the data depth function

ABSTRACT

The clustering of catchments is important for prediction in ungauged basins, model parameterization and watershed development and management. The aim of this study is to explore a new measure of similarity among catchments, using a data depth function and comparing it with catchment clustering indices based on flow and physical characteristics. A cluster analysis was performed for each similarity measure using the affinity propagation clustering algorithm. We evaluated the similarity measure based on depth–depth plots (DD-plots) as a basis for transferring parameter sets of a hydrological model between catchments. A case study was developed with 21 catchments in a diverse New Zealand region. Results show that clustering based on the depth–depth measure is dissimilar to clustering on catchment characteristics, flow, or flow indices. A hydrological model was calibrated for the 21 catchments and the transferability of model parameters among similar catchments was tested within and between clusters defined by each clustering method. The mean model performance for parameters transferred within a group always outperformed those from outside the group. The DD-plot based method was found to produce the best in-group performance and second-highest difference between in-group and out-group performance.

EDITOR D. Koutsoyiannis; ASSOCIATE EDITOR A. Viglione     

Risk assessment using suprema data

This paper proposes a stochastic approach to model temperature dynamic and study related risk measures. The dynamic of temperatures can be modelled by a mean-reverting process such as an Ornstein–Uhlenbeck one. In this study, we estimate the parameters of this process thanks to daily observed suprema of temperatures, which are the only data gathered by some weather stations. The expression of the cumulative distribution function of the supremum is obtained thanks to the law of the hitting time. The parameters are estimated by a least square method quantiles based on this function. Theoretical results, including mixing property and consistency of model parameters estimation, are provided. The parameters estimation is assessed on simulated data and performed on real ones. Numerical illustrations are given for both data. This estimation will allow us to estimate risk measures, such as the probability of heat wave and the mean duration of an heat wave.     

Snow water equivalent prediction using Bayesian data assimilation methods

Using the U.S. National Weather Service’s SNOW-17 model, this study compares common sequential data assimilation methods, the ensemble Kalman filter (EnKF), the ensemble square root filter (EnSRF), and four variants of the particle filter (PF), to predict seasonal snow water equivalent (SWE) within a small watershed near Lake Tahoe, California. In addition to SWE estimation, the various data assimilation methods are used to estimate five of the most sensitive parameters of SNOW-17 by allowing them to evolve with the dynamical system. Unlike Kalman filters, particle filters do not require Gaussian assumptions for the posterior distribution of the state variables. However, the likelihood function used to scale particle weights is often assumed to be Gaussian. This study evaluates the use of an empirical cumulative distribution function (ECDF) based on the Kaplan–Meier survival probability method to compute particle weights. These weights are then used in different particle filter resampling schemes. Detailed analyses are conducted for synthetic and real data assimilation and an assessment of the procedures is made. The results suggest that the particle filter, especially the empirical likelihood variant, is superior to the ensemble Kalman filter based methods for predicting model states, as well as model parameters.     

Functional data clustering using K-means and random projection with applications to climatological data

In this paper, an efficient pattern recognition method for functional data is introduced. The proposed method works based on reproducing kernel Hilbert space (RKHS), random projection and K-means algorithm. First, the infinite dimensional data are projected onto RKHS, then they are projected iteratively onto some spaces with increasing dimension via random projection. K-means algorithm is applied to the projected data, and its solution is used to start K-means on the projected data in the next spaces. We implement the proposed algorithm on some simulated and climatological datasets and compare the obtained results with those achieved by K-means clustering using a single random projection and classical K-means. The proposed algorithm presents better results based on mean square distance (MSD) and Rand index as we have expected. Furthermore, a new kernel based on a wavelet function is used that gives a suitable reconstruction of curves, and the results are satisfactory.     

Nonlinear inversion of electrical resistivity imaging using pruning Bayesian neural networks

Conventional artificial neural networks used to solve electrical resistivity imaging (ERI) inversion problem suffer from overfitting and local minima. To solve these problems, we propose to use a pruning Bayesian neural network (PBNN) nonlinear inversion method and a sample design method based on the K-medoids clustering algorithm. In the sample design method, the training samples of the neural network are designed according to the prior information provided by the K-medoids clustering results; thus, the training process of the neural network is well guided. The proposed PBNN, based on Bayesian regularization, is used to select the hidden layer structure by assessing the effect of each hidden neuron to the inversion results. Then, the hyperparameter α _k , which is based on the generalized mean, is chosen to guide the pruning process according to the prior distribution of the training samples under the small-sample condition. The proposed algorithm is more efficient than other common adaptive regularization methods in geophysics. The inversion of synthetic data and field data suggests that the proposed method suppresses the noise in the neural network training stage and enhances the generalization. The inversion results with the proposed method are better than those of the BPNN, RBFNN, and RRBFNN inversion methods as well as the conventional least squares inversion.     

3D mapping of buried underworld infrastructure using dynamic Bayesian network based multi-sensory image data fusion

The successful operation of buried infrastructure within urban environments is fundamental to the conservation of modern living standards. In this paper a novel multi-sensor image fusion framework has been proposed and investigated using dynamic Bayesian network for automatic detection of buried underworld infrastructure. Experimental multi-sensors images were acquired for a known buried plastic water pipe using Vibro-acoustic sensor based location methods and Ground Penetrating Radar imaging system. Computationally intelligent conventional image processing techniques were used to process three types of sensory images. Independently extracted depth and location information from different images regarding the target pipe were fused together using dynamic Bayesian network to predict the maximum probable location and depth of the pipe. The outcome from this study was very encouraging as it was able to detect the target pipe with high accuracy compared with the currently existing pipe survey map. The approach was also applied successfully to produce a best probable 3D buried asset map.     

微地震资料贝叶斯理论差分进化反演方法

微地震监测难以拾取准确初至,为了提高反演定位精度和减小多解性,研究了微地震贝叶斯差分进化反演方法.从分析讨论理论模型反演残差及其协方差分布特征出发,结合对比不加噪音和加入不同程度的噪音后残差协方差极小点位置移动、分布梯度变化特征,提出了先验信息解估计方法.针对后验估计中,由于难以获得先验信息解的方差估计致使无法计算加权系数问题,通过分析残差变化特征和解的变化关系,研究了利用残差求取加权系数的方法.为了加快寻优速度,讨论了差分进化反演方法,在变异操作方面使用差分策略,即利用种群中个体间的差分向量对个体进行扰动,实现个体变异,充分有效利用群体分布特性,提高算法的搜索能力,避免遗传算法中变异方式的不足.通过理论模型测试本方法的反演效果,并且和搜索方法反结果进行比较.测试结果证明本反演方法,对于不同程度初至干扰,反演结果向准确解逼近程度比搜索方法要好得多,实际资料的反演结果也好于搜索方法.     

Inversion of nuclear well-logging data using neural networks

This work looks at the application of neural networks in geophysical well‐logging problems and specifically their utilization for inversion of nuclear downhole data. Simulated neutron and γ‐ray fluxes at a given detector location within a neutron logging tool were inverted to obtain formation properties such as porosity, salinity and oil/water saturation. To achieve this, the forward particle‐radiation transport problem was first solved for different energy groups (47 neutron groups and 20 γ‐ray groups) using the multigroup code EVENT. A neural network for each of the neutron and γ‐ray energy groups was trained to re‐produce the detector fluxes using the forward modelling results from 504 scenarios. The networks were subsequently tested on unseen data sets and the unseen input parameters (formation properties) were then predicted using a global search procedure. The results obtained are very encouraging with formation properties being predicted to within 10% average relative error. The examples presented show that neural networks can be applied successfully to nuclear well‐logging problems. This enables the implementation of a fast inversion procedure, yielding quick and reliable values for unknown subsurface properties such as porosity, salinity and oil saturation.     

Inversion of DC resistivity data using neural networks

The inversion of geoelectrical resistivity data is a difficult task due to its non-linear nature. In this work, the neural network (NN) approach is studied to solve both 1D and 2D resistivity inverse problems. The efficiency of a widespread, supervised training network, the back-propagation technique and its applicability to the resistivity problem, is investigated. Several NN paradigms have been tried on a basis of trial-and-error for two types of data set. In the 1D problem, the batch back-propagation paradigm was efficient while another paradigm, called resilient propagation, was used in the 2D problem. The network was trained with synthetic examples and tested on another set of synthetic data as well as on the field data. The neural network gave a result highly correlated with that of conventional serial algorithms. It proved to be a fast, accurate and objective method for depth and resistivity estimation of both 1D and 2D DC resistivity data. The main advantage of using NN for resistivity inversion is that once the network has been trained it can perform the inversion of any vertical electrical sounding data set very rapidly.     

Bayesian spectral density approach for modal updating using ambient data

The problem of identification of the modal parameters of a structural model using measured ambient response time histories is addressed. A Bayesian spectral density approach (BSDA) for modal updating is presented which uses the statistical properties of a spectral density estimator to obtain not only the optimal values of the updated modal parameters but also their associated uncertainties by calculating the posterior joint probability distribution of these parameters. Calculation of the uncertainties of the identified modal parameters is very important if one plans to proceed with the updating of a theoretical finite element model based on modal estimates. It is found that the updated PDF of the modal parameters can be well approximated by a Gaussian distribution centred at the optimal parameters at which the posterior PDF is maximized. Examples using simulated data are presented to illustrate the proposed method. Copyright © 2001 John Wiley & Sons, Ltd.     

Seismic hazard assessment using intensity point data

The influence of the attenuation model used in seismic hazard assessment in terms of intensity and acceleration is studied. For two sites in central Italy, the catalogue of the actual observed intensities during the last three centuries has been recovered. In the study region, the data collected during a recent seismic sequence give the basis for relating intensity and acceleration. The results show the importance of establishing statistical relationships among the used quantities, based on a representative set of data.     

Ground water quality assessment using multi-rectangular diagrams

A new graphical technique is proposed here for classifying chemical analyses of ground water. In this technique, a diagram is constructed using rectangular coordinates. The new diagram, called a multi-rectangular diagram (MRD), uses adjacent multi-rectangles in which each rectangle represents a specific ground water type. This new diagram has the capability to accommodate a large number of data sets. MRDs have been used to classify chemical analyses of ground water in the Chaj Doab area of Pakistan to illustrate this new approach. Using this graphical method, the differentiated ground water types are calcium bicarbonate, magnesium bicarbonate, sodium bicarbonate, and sodium sulfate. Sodium bicarbonate emerges as the most abundant ground water type. MRDs also offer a visual display of the Chebotarev sequence of ground water quality evolution.     

Seismic reliability assessment of lifeline networks using clustering‐based multi‐scale approach

Seismic reliability assessment of lifeline networks gives rise to various technical challenges, which are mostly caused by a large number of network components, complex network topology, and statistical dependence between component failures. For effective risk assessment and probabilistic inference based on post‐hazard observations, various non‐simulation‐based algorithms have been developed, including the selective recursive decomposition algorithm (S‐RDA). To facilitate the application of such an algorithm to large networks, a new multi‐scale approach is developed in this paper. Using spectral clustering algorithms, a network is first divided into an adequate number of clusters such that the number of inter‐cluster links is minimized while the number of the nodes in each cluster remains reasonably large. The connectivity around the identified clusters is represented by super‐links. The reduced size of the simplified network enables the S‐RDA algorithm to perform the network risk assessment efficiently. When the simplified network is still large even after a clustering, additional levels of clustering can be introduced to have a hierarchical modeling structure. The efficiency and effectiveness of the proposed multi‐scale approach are demonstrated successfully by numerical examples of a hypothetical network, a gas transmission pipeline network, and a water transmission network. Copyright © 2014 John Wiley & Sons, Ltd.     

An assessment of large-scale flood modelling based on LiDAR data

Large-scale flood modelling approaches designed for regional to continental scales usually rely on relatively simple assumptions to represent the potentially highly complex river bathymetry at the watershed scale based on digital elevation models (DEMs) with a resolution in the range of 25–30 m. Here, high-resolution (1 m) LiDAR DEMs are employed to present a novel large-scale methodology using a more realistic estimation of bathymetry based on hydrogeomorphological GIS tools to extract water surface slope. The large-scale 1D/2D flood model LISFLOOD-FP is applied to validate the simulated flood levels using detailed water level data in four different watersheds in Quebec (Canada), including continuous profiles over extensive distances measured with the HydroBall technology. A GIS-automated procedure allows to obtain the average width required to run LISFLOOD-FP. The GIS-automated procedure to estimate bathymetry from LiDAR water surface data uses a hydraulic inverse problem based on discharge at the time of acquisition of LiDAR data. A tiling approach, allowing several small independent hydraulic simulations to cover an entire watershed, greatly improves processing time to simulate large watersheds with a 10-m resampled LiDAR DEM. Results show significant improvements to large-scale flood modelling at the watershed scale with standard deviation in the range of 0.30 m and an average fit of around 90%. The main advantage of the proposed approach is to avoid the need to collect expensive bathymetry data to efficiently and accurately simulate flood levels over extensive areas.