Deep neural network application for 4D seismic inversion to changes in pressure and saturation: Optimizing the use of synthetic training datasets

In this work, we tackle the challenge of quantitative estimation of reservoir dynamic property variations during a period of production, directly from four-dimensional seismic data in the amplitude domain. We employ a deep neural network to invert four-dimensional seismic amplitude maps to the simultaneous changes in pressure, water and gas saturations. The method is applied to a real field data case, where, as is common in such applications, the data measured at the wells are insufficient for properly training deep neural networks, thus, the network is trained on synthetic data. Training on synthetic data offers much freedom in designing a training dataset, therefore, it is important to understand the impact of the data distribution on the inversion results. To define the best way to construct a synthetic training dataset, we perform a study on four different approaches to populating the training set making remarks on data sizes, network generality and the impact of physics-based constraints. Using the results of a reservoir simulation model to populate our training datasets, we demonstrate the benefits of restricting training samples to fluid flow consistent combinations in the dynamic reservoir property domain. With this the network learns the physical correlations present in the training set, incorporating this information into the inference process, which allows it to make inferences on properties to which the seismic data are most uncertain. Additionally, we demonstrate the importance of applying regularization techniques such as adding noise to the synthetic data for training and show a possibility of estimating uncertainties in the inversion results by training multiple networks.     

Water quality monitoring at a virtual watershed monitoring station using a modified deep extreme learning machine

ABSTRACT

A new deep extreme learning machine (ELM) model is developed to predict water temperature and conductivity at a virtual monitoring station. Based on previous research, a modified ELM auto-encoder is developed to extract more robust invariance among the water quality data. A weighted ELM that takes seasonal variation as the basis of weighting is used to predict the actual value of water quality parameters at sites which only have historical data and no longer generate new data. The performance of the proposed model is validated against the monthly data from eight monitoring stations on the Zengwen River, Taiwan (2002–2017). Based on root mean square error, mean absolute error, mean absolute percentage error and correlation coefficient, the experimental results show that the new model is better than the other classical spatial interpolation methods.     

长江中下游湖泊群不同生活型水生植物分布的时空变化(1986—2020年)

不同生活型水生植物对水环境的影响和碳固持能力不同,开展大尺度范围内不同生活型水生植物的时空分布和动态变化研究,是全面掌握湖泊水生态环境变化趋势、准确核算水生生态系统碳源/碳汇的前提。以长江中下游10 km²以上(共131个)的湖泊为研究对象,基于野外调查和先验知识,通过光谱分析,研发了不同生活型水生植物遥感高精度机器学习识别算法,解析了长江中下游湖泊群不同生活型水生植物的时空变化规律。研究表明,长江中下游湖泊群不同生活型水生植物遥感监测精度为0.81,Kappa系数为0.74;1986—2020年长江中下游湖泊群水生植物面积为2541.58~4571.42 km²,占湖泊总面积的15.99%~28.77%,沉水植物是优势类型(Max_1995年=2649.21 km²,Min_2005年=921.38 km²),其次是挺水植物(Max_2005年=1779.44 km²,Min_2020年=569.05 km²)和浮叶植物(Max_2015年=685.68 km²,Min_2000年=293.04 km²);水生植物主要分布在长江干流流域湖泊群,其次是鄱阳湖流域、洞庭湖流域、太湖流域和汉江流域;变化趋势上,1986—2020年长江中下游湖泊群水生植物面积呈现先增长(1986—1995年)、后下降(1995—2010年)、再增加(2010年后)的趋势。本研究可为长江中下游湖泊群生态环境调查及水环境管理提供重要参考。     

Stripe-based fragility analysis of multispan concrete bridge classes using machine learning techniques

A framework for the generation of bridge-specific fragility curves utilizing the capabilities of machine learning and stripe-based approach is presented in this paper. The proposed methodology using random forests helps to generate or update fragility curves for a new set of input parameters with less computational effort and expensive resimulation. The methodology does not place any assumptions on the demand model of various components and helps to identify the relative importance of each uncertain variable in their seismic demand model. The methodology is demonstrated through the case study of a multispan concrete bridge class in California. Geometric, material, and structural uncertainties are accounted for in the generation of bridge numerical models and their fragility curves. It is also noted that the traditional lognormality assumption on the demand model leads to unrealistic fragility estimates. Fragility results obtained by the proposed methodology can be deployed in a risk assessment platform such as HAZUS for regional loss estimation.     

高层建筑坍塌墙体修复施工安全管理定量化控制模型研究

为了确保震后高层建筑坍塌墙体修复工程的安全性,提出一种高层建筑震后坍塌墙体修复施工安全管理方法。结合主成分分析法与信息熵法对震后高层建筑坍塌墙体修复施工管理的控制特征量进行提取,确定模型评价指标权重;根据加权学习法对坍塌墙体修复的施工进行量化控制;通过灰阶量化评估进行约束参量分析,构建安全管理控制模型;采用管理因素模糊调度算法实现施工安全管理控制,并完成优化决策。通过仿真实验验证坍塌墙体修复施工管理控制的管理效益与累积评价百分率。分析结果表明,采用该方法进行震后高层建筑坍塌墙体修复施工管理控制的管理效益最高可达83.32%,累积管理效果回报率较高,累积评价百分率平均约为75%,能够有效提高施工过程的安全管理效能,确保施工安全。     

基于多尺度特征深度学习的短临降水预报

雷达回波外推是短临降水预报的一种重要方法。针对雷达回波外推中随着外推时间的增加而出现回波演变信息丢失这一问题,本文提出一种多尺度特征融合的深度学习短临降水预报模型(以下简称为MSF2)。首先,采用多尺度的卷积核对网络的浅层信息进行特征提取,弥补单一特征检测带来的不足。其次,将不同维度的特征信息进行拼接及通道混洗,进一步增强特征图通道之间的信息流通和信息表达能力。最后,将特征图中的多尺度信息进行融合,从而有效保留不同尺度的特征信息。利用华南雷达回波拼图资料数据,在3种不同降水强度（5 mm/h、10 mm/h和25 mm/h）下进行降水预报研究,并与光流法和ConvLSTM两种主流算法进行了对比。结果显示,在3种不同降水强度条件下,MSF2在所有评价指标（命中率POD、临界成功指数CSI、误报率FAR）中表现最优,这表明引入多尺度机制能改善模型的特征提取能力。相比于目前主流的光流法和ConvLSTM,本文提出的模型对于短临降水预报具有较好的适用性和较高的预报精度,而且实现了业务化运行。     

基于深度学习和时空特征融合的海洋渔船密度预测方法

为了从海量渔船轨迹数据中挖掘隐含的信息和知识,进而为渔业行政主管部门的决策提供科学依据,本研究以AIS渔船轨迹数据为研究对象,提出了一种基于深度学习和面向时空特征融合的海洋渔船密度预测方法：首先,利用渔船轨迹数据集对渔船行驶区域进行网格划分;其次,筛选出渔船高密度区域进行研究,避免数据稀疏性问题;再次,根据渔船轨迹数据的时空分析,构建三维时空融合矩阵;最后,通过卷积循环神经网络模型捕获渔船分布的时间和空间特征,并利用卷积神经网络的堆叠加强对空间特征的学习。实验通过东海海域渔船真实轨迹数据进行具体测试,结果表明渔船密度预测值与真实值非常接近,平均绝对误差为4×10-4,模型较好地拟合了渔船密度分布特征,有效地提高了渔船捕捞热点预测的准确性和鲁棒性。     

基于极限学习机的GF-2影像分类

遥感图像分类是提取图像有效信息过程中重要的一部分,为了探寻最优的分类方法,许多机器学习算法逐步应用于遥感分类中。极限学习机（extreme learning machine,ELM）以其高效、快速和良好的泛化性能在模式识别领域得到广泛应用。本文采用训练速度快、运算量小的极限学习机算法与支持向量机（support vector machines,SVM）算法和最大似然法进行分类对比,对高分辨率遥感图像进行分类,分析极限学习机算法对于遥感图像分类的准确度等性能。选取吉林省长春市部分区域的GF-2遥感数据,将融合后的影像设置为原始数据,利用3种方法进行分类。研究结果表明,极限学习机算法分类图像总体分类精度达到85%以上,kappa系数达到0.718,与其他分类方法相比分类准确度较高,且极限学习机运行时间比支持向量机运行时间约短2 480 s,约为支持向量机运行时间的1/8,因此具有良好的性能和实用价值。     

基于年际增量法的卷积神经网络模型对中国夏季降水的预测研究

将前冬的500 hPa位势高度、向外长波辐射和海表温度的年际增量作为预测因子,建立基于卷积神经网络(Convolutional Neural Network, CNN)的非线性预测模型,对中国160个测站夏季降水展开预测研究,并与基于线性奇异值分解(Singular Value Decomposition, SVD)的预测模型进行效果对比。结果表明：CNN在1981—2020年的交叉检验中所回报的降水平均PS评分和距平相关系数(ACC)分别为74.33和0.12,比SVD高2.15和0.06,说明CNN比SVD在整体上对夏季降水具有更好的预测能力。其中,CNN对SVD预测较好年份的预测效果提升较为明显,对SVD预测较差的年份则改进不大。CNN对中国降水预测存在一定的系统性偏差,订正后CNN对拉尼娜年的降水预测改进较大。结果表明,基于年际增量法的CNN预测模型展示出较好的潜在应用价值。     

Satoumi concept illustrated by sustainable bottom-up initiatives of Japanese Fisheries Cooperative Associations

“Satoumi” is a Japanese term meaning a seascape where human-ecosystem interaction has resulted in increased biodiversity and productivity, thus improving the health of the environment and its ecosystem services. However, this definition does not do justice to the universality of the concept, and the holistic ideas and related valuable practices promoted by the term, applicable to fisheries science, contributing thus to the neglect of the term in the international stage. This article explores the importance of environmental, commercial, and research activities related to Satoumi, based on actions implemented by Fisheries Cooperative Associations (FCAs) of different prefectures in Japan. The examined case studies are bottom-up initiatives undertaken by Japanese fishermen and constitute effective means towards sustainable and competitive fisheries management. Such initiatives play an important role in disseminating the positive effects of the Satoumi approach in Japan, and promote therefore its adoption in other regions of the world as well.