基于数据分布域变换与贝叶斯神经网络的渗透率预测及不确定性估计

渗透率是储层评价和油气藏开发的关键参数.传统测井方法与常规机器学习方法估算的渗透率都是固定值.但由于测井数据本身存在噪声, 渗透率的预测结果可能受到噪声的影响出现测量性的随机误差(即任意不确定性); 同时, 当测试数据与训练数据存在差异时, 机器学习模型在预测渗透率时可能出现模型参数的不确定性(即认知不确定性).为实现渗透率的准确预测并量化两种不确定性对结果的影响, 本文提出基于数据分布域变换和贝叶斯神经网络同时实现渗透率预测及其不确定性的估计.提出方法主要包括两个部分: 一部分是不同域数据分布的相互转换, 另一部分是基于贝叶斯理论的神经网络渗透率建模预测和不确定性估计.由于贝叶斯神经网络存在数据分布的假设, 当标签的概率分布与网络的分布保持一致时, 贝叶斯神经网络可以更好的学习到数据之间的关系.因此通过寻找一个函数将一个原始域的渗透率标签转换为目标域的与渗透率有关的变量(我们称为目标域渗透率), 使得该变量符合贝叶斯神经网络的分布假设.我们使用贝叶斯神经网络预测目标域渗透率以及任意不确定性和认知不确定性.随后, 通过分布域的逆变换, 我们将目标域渗透率还原回原始域渗透率.应用本文方法到某油田的18口井的测井数据中, 使用16口井的数据进行训练, 2口井进行测试.测试井的预测渗透率与真实渗透率基本一致.同时, 任意不确定性的预测结果提供了渗透率预测值受到的测井数据噪声影响的位置.认知不确定的预测结果说明数据量少的位置具有更高的认知不确定性.我们提出的这一流程不仅显示了在储层表征方面的巨大潜力, 同时可以降低测井解释时的风险.

     

地震观测仪器自动监控软件设计与实现

结合地震地球物理观测台网台站监测工作实际,研制一款地震观测仪器自动监控软件。该软件适用于不同网络环境下IP类仪器设备的网络实时或定时自动监控,地球物理观测类仪器时钟、状态和观测数据的定时自动监控,能自动识别当日观测数据中的无效数据、变幅异常、干扰、地震和限幅等信息,无需借助其他硬件和软件即可将告警信息以短信方式发送给值班人员和管理人员,可实现对仪器原始观测数据的自动备份和手动恢复,及对仪器的一键重启和校正时钟等控制类操作,无需跨平台操作即可查看当日观测数据曲线及分析数据的完整性和有效性,从而实现对仪器的统一高效管理。     

中国水质站网规划中几个问题的探讨

在完成中国水质站网规划工作的基础上,对下列问题作了进一步探讨:(1)水质站网规划单元的划分;(2)聚类分析在水质站网规划中的应用;(3)站网密度与人口和经济发展的关系;(4)中国水质站网的分区特征。     

深海底リアルタイム観測ネットワークの構築

近年来,我国地震观测取得了长足进展,同时海底观测系统的建设也方兴未艾。而欧美及日本等发达国家在海底有线实时监测的成功先例,为我国相应工作的开展提供了良好的参考及借鉴。其中日本海底有线实时监测系统DONET始建于2011年,专注于海底地动—水压信号的高效采集,旨在监视日本南海海槽的地震和海啸事件。本文主要介绍DONET的硬件布局、搭建流程以及科研产出情况,并简要介绍我国国家海底科学观测网的基本信息。     

Modelling combined open channel flow by artificial neural networks

Combined open channel flow is encountered in many hydraulic engineering structures and processes, such as irrigation ditches and wastewater treatment facilities. Extensive experimental studies have conducted to investigate combined flow characteristics. Nevertheless, there is no simple relationship that can fully describe the velocity profiles in a turbulent flow. The artificial neural network (ANN) has great computational capability for solving various complex problems, such as function approximation. The main objective of this study is to evaluate the applicability of the ANN for simulating velocity profiles, velocity contours and estimating the discharges accordingly. The velocity profiles measured by an acoustic doppler velocimeter in the open channel of the Chihtan purification plant, Taipei, with different discharges at fixed measuring section and different depths are presented. The total number of data sets is 640 and the data sets are split into two subsets, i.e. training and validation sets. The backpropagation algorithm is used to construct the neural network. The results demonstrate that the velocity profiles can be modelled by the ANN, and the ANN constructed can nicely fit the velocity profiles and can precisely predict the discharges for the conditions investigated. Copyright © 2005 John Wiley & Sons, Ltd.     

Approaches for modeling longitudinal dispersion in pore-networks

The purpose of this study is to quantify the dispersivity in the longitudinal direction by upscaling pore scale mixing over a network domain and to verify the dispersivity with that obtained through the more rigorous upscaling technique, the Brownian particle tracking model (BPTM). We model a porous medium with a network of pore-units that are comprised of pore bodies and bonds of finite volume. Such a pore-unit is assumed to be a mixing cell with the steady state flow condition for a single fluid. Dispersivity can be obtained by solving the mixing cell model (MCM) for the concentration in each pore-unit and by averaging the concentrations for a large number of pore units (as a function of time and space). A minimal size of network that ascertains an asymptotic value of dispersivity was determined and verified with large size pore networks. This numerically computed dispersivity is compared with the results from the BPTM for the same porous medium and flow conditions. We show that the dispersivity obtained from the MCM is equally reliable for the heterogeneous pore-networks and can be estimated as a function of pore size heterogeneity. For homogeneous networks with the MCM, the iteration time step plays an important role. On the other hand, for networks with the BPTM, the assumption of intra-bond velocity profile affects the results.     

Location Accuracy of the China National Seismograph Network Estimated by Repeating Events

INTRODUCTIONDigital seismological observation in China has had a significant development in recent years,especiallysince the last five years(Liu Ruifeng,et al.,2003).For further development,it isnecessarytoassessthe monitoringcapabilityof the existingseismological network.One of theimportantassessments is the estimation of regionalizedlocation accuracy.Upto present,several approaches have been proposed to assess the location accuracy,such asthe groundtruth event approach(e.g.,Lienert,199…     

X射线荧光方法进行野外找矿及成矿规律研究的现状和前景

目前,能量色散Ｘ射线荧光现场测量方法已经逐步完善,能够在现场取得多种元素含量的近似定量数据,为地质勘查工作提供元素含量及其分布的信息,在此基础上要求加大方法应用的深度和广度,将Ｘ射线荧光方法取得的资料应用于矿产预测,成矿规律,地质构造及基础地质研究等方面,本文根据作者近年的实际工作成果并汇集有关资料,用丰富的实例文章爱一领域的现状,并提出今后发展的重点和前景。     

Stroke构造、移位一体化的道路网示意化方法

针对当前常用示意性地图生成方法中往往存在简化程度不够、未考虑长度信息以及时间效率低等问题,提出了stroke构造、移位一体化的道路网示意化方法。该方法在对道路网进行stroke划分构造的同时,直接对其进行渐进式的移位示意化及拓扑检查。同时本文还提出利用相似分形维数来定量比较并验证不同示意化方法的有效性。试验表明,本算法在考虑原始线形的基础上,示意化过程简单,时间效率较高,减少了拓扑冲突问题,保证了拓扑一致性及路网的均衡分布,具有较好的清晰度和认知度。     

无人机三维倾斜摄影测量助力自然资源调查——以不动产调查为例

随着第一次地理国情普查与监测、第三次全国土地调查等重大自然资源调查类项目的实施,以无人机作为搭载平台的遥感影像获取手段在不断发展并深入应用,特别是无人机航测技术和地理信息技术的结合应用越来越紧密.本文以不动产调查为例,重点论述结合无人机三维倾斜摄影测量技术在内的测绘地理信息新手段的作业原理及应用情况,并对其优缺点进行总...