基于深度学习到时拾取自动构建长宁地震前震目录

将深度学习到时拾取、震相关联技术与传统定位方法联系起来,构建一套连续波形自动化处理与地震目录自动构建流程,对于高效充分利用地震资料,提升微震检测能力具有十分重要的意义.我们应用最新发展的迁移学习震相识别技术、震相自动关联技术,对长宁Ms6.0地震震中附近21个台站震前半个月(6月1日-6月17日)的连续记录波形进行P、...     

Countermeasures for breakwater foundation subjected to foreshocks and main shock of earthquake loading

Coastal protective structures, such as composite breakwaters, are generally vulnerable to earthquake. It was observed that breakwaters damage mainly due to failure of their foundations. However, the seismically induced failure process of breakwater foundation has not been well understood. This study describes failure mechanism of breakwater foundation as well as a newly developed reinforcing model for breakwater foundation that can render resiliency to breakwater against earthquake-related disasters. Steel sheet piles and gabions were used as reinforcing materials for foundation. The experimental program consisted of a series of shaking table tests for conventional and reinforced foundation of breakwater. Numerical analyses were conducted using finite difference method, and it was observed that the numerical models were capable to elucidate the seismic behavior of soil–reinforcement–breakwater system. This paper presents an overview of the results of experimental and numerical studies of the seismic response of breakwater foundation. Overall, the results of these studies show the effectiveness of the reinforced foundation in mitigating the earthquake-induced damage to the breakwater. Moreover, numerical simulation was used for parametric study to determine the effect of different embedment depths of sheet piles on the performance of breakwater foundation subjected to seismic loading.     

基于迁移学习方法的山东冷云冰晶特征分布研究

为分析冷云中冰晶的分布特征,揭示冰晶增长演变机制,根据冰晶形状和尺度特征分为8类并进行标注,同时标注1类隔断栏进行数据质量控制,将9类标签图像整合并建立图像集,利用迁移学习VGGNet16方法进行识别训练,经训练模型分类准确率达98%。将模型应用到秋季冷云冰晶特征研究中,选取3次积层混合云和3次层状云降水过程,分析冰晶形状在不同温度区间的占比及冰晶谱变化特征,结果表明,温度通过影响冰晶基面与棱面的比值来决定冰晶初始形状分布,相同温度区间积层混合云内球状冰晶和线型冰晶占比高于层状云,低于-12℃后各类冰晶占比相对固定;积层混合云内线型冰晶直径集中在300~800 μm,冰晶谱呈多峰分布,聚合体直径大于600 μm,冰晶谱首尾两端浓度相当,球状冰晶直径集中在120～300 μm,冰晶谱呈单调下降趋势。     

基于注意力机制与自适应时序分解的气温预报模型

针对传统方法在捕捉气象序列长期依赖关系及泛化性能上的不足,提出了一种基于稀疏注意力与自适应时序分解的气温预报模型（ATFSAS）。该模型整体采用编码器 解码器架构,结合稀疏注意力机制以有效捕捉气象观测数据间的长期依赖性。为减少编码过程中造成的冗余,提出了一种信息蒸馏方法。通过结合多层解码器与自适应时序分解单元,逐步细化预报信号中的周期性和趋势性分量,实现了较为精准的气温预报。基于德国耶拿气象数据集,进行24 h精细化气温预报,其平均绝对误差为1.7108℃。基于中国地面气候资料日值数据集,进行中短期日平均气温预报和多地区单日平均气温预报,相比传统模型LSTM,ATFSAS模型预报结果的平均绝对误差分别提升了35.56%和23.66%。     

基于机器学习的格点气温预报订正方法

使用2017年9月至2021年3月国家级业务化运行的智能网格实况分析产品和欧洲中期天气预报中心全球模式(EC)产品,根据湖北省的地理分布特征构建6个分区,采用基于LightGBM机器学习算法建立的气温预报方法,生成湖北省0.05°×0.05°格点气温预报产品。利用2021年4—9月的预报产品和格点实况资料进行检验,结果表明:基于机器学习的气温预报方法(MLT)取得了较好的预报效果,其在0～72 h时效内优于中央气象台下发的气温精细化指导预报(SCMOC)和EC产品;MLT在山区的误差较平原大,但山区的订正幅度大于平原,日最高气温的订正幅度大于日最低气温的订正幅度;4—9月MLT、SCMOC、EC产品的平均绝对误差(MAE)日变化都呈现了白天偏高、夜间偏低、午后凸起的单峰特征,MLT的MAE值较SCMOC和EC产品的更低,并且在转折性天气中仍具有优势;站点检验与格点检验结论一致,基于格点建模的气温预报产品对站点预报同样得到了订正。机器学习在格点气温的模式订正方面可以作为一个行之有效的手段。     

遥感影像分析在新增建设用地督察中的应用

以卫星影像大数据和时空数据挖掘技术为基础,采用深度学习框架支撑,构建了面向新增建设用地遥感监测应用需求的深度学习训练平台,可智能发现自然资源督察线索,全面提高自然资源督察问题的发现能力.该技术可广泛服务于大范围、高频次的督察工作,为督察工作提供更及时客观全面的信息基础和保障.     

Seismic stability analysis of piled embankments: A multiphase approach

The seismic stability analysis of an embankment lying over a soft foundation soil reinforced by a group of vertical piles is performed within the framework of the upper bound kinematic approach of yield design. The analysis is based on a previously developed ‘multiphase’ model of the reinforced ground, which explicitly accounts for the shear and bending resistances of the piles. Making use of appropriate failure mechanisms involving shear zones across which the reinforcements are continuously deforming, along with ‘plastic hinge’ surfaces, upper bound estimates to the critical seismic coefficient of the structure are derived. The results, which are confirmed by the simulations obtained from a finite element elastoplastic code, give clear evidence of the key role played by the bending strength capacities of the piles in ensuring the stability of the pile reinforced embankment. Copyright © 2009 John Wiley & Sons, Ltd.     

粉煤灰碎石混凝土灌注桩法加固软弱地基效果的分析与检验

对由深厚软弱土层组成的地基,采用粉煤灰碎石混凝土灌注桩法予以加固处理时,其施工机具相对简单,工期短,节约“三材”,投资省,具有显著的经济效益。由粉煤灰碎石混凝土灌注桩和深厚软弱土层共同构成的复合地基的工程特性,介于一般碎石桩复合地基与混凝土桩桩基之间,其基本特征是:承载力的提高,主要是置于深厚软弱土层中的粉煤灰碎石混凝土灌注桩的排水和垂直加筋作用的结果;其承载机理,部分具有一般碎石桩复合地基的特征;在结构物荷载作用下,其桩、土荷载分担特征,比较接近于普通混凝土桩桩基。     

Comparing Human and Neural Network Learning of Climate Categories

This study compares how humans and neural networks classify climate types. Human subjects were asked to classify climates from monthly temperature and precipitation patterns. To model their learning process, the same data were used to produce input vectors that trained a pattern associator neural network. Both human subjects and the neural network classified climates accurately after 10 rounds of supervised learning. The neural network successfully modeled the rate of human learning and the ability to learn specific climate categories. Moreover, the neural network weights used to classify climates correspond to distinct visual characteristics in temperature and precipitation. These results suggest that neural networks can model the formation of visual categories.     

On the enhanced confining pressure aproach to the mechanics of reinforced soil

Summary A review is presented of conceptual approaches that are currently in use for interpreting the operation of reinforced soil. The concepts of enhanced confining pressure and of reduced normal tensile strains are found to be closely related to current experimental and theoretical investigations as well as to the design methodologies available at present for reinforced soil applications. Focusing on the enhanced confining pressure concept, analytical expressions are developed that allow the estimation of the value of equivalent confining stress increase when the properties of soil and reinforcement and the state of stresses acting on a cylindrical reinforced soil element are known. The derived expressions can also be utilized for estimating the soil-reinforcement friction angle from the results of triaxial tests on cylindrical samples reinforced with horizontal layers of reinforcement.