融合浅层特征的深度卷积神经网络互花米草遥感监测方法     

朱玉玲  王建步  王安东  王锦锦  赵晓龙  任广波  胡亚斌  陈晓英  马毅 《海洋科学》2019,43(7):12-22

基于2018年10月份黄河口入海两侧的LANDSAT8 OLI影像,提取植被指数和缨帽变换分量共9维光谱特征,构建融合浅层特征的8层深度卷积神经网络(deepconvolutionalneuralnetwork,DCNN)分类模型,开展互花米草(SpartinaalternifloraLoisel)遥感监测的方法研究,并从不同的浅层特征来具体分析互花米草的监测结果。结果表明:(1)在分类方法上,DCNN模型的总体分类精度最高,达到90.33%,与支持向量机(support vector machine, SVM)、随机森林(random forest, RF)分类器相比,精度分别提高4.78%、2.7%,互花米草的生产者精度分别提高了2.56%、0.47%,说明在滨海湿地遥感影像分类中, DCNN有着更好的应用潜力;(2)融合浅层特征后, DCNN的总体分类精度和互花米草的识别精度分别提高了0.34%和3.25%,有效提高了对互花米草的监测能力。其中,融合归一化植被水分指数(NDII)浅层特征的DCNN分类方法中,互花米草的识别精度提高最多,为2.56%,比值植被指数(RVI)次之,为2.32%。研究结果可为互花米草的监测与管理提供技术与数据支撑。  相似文献   

基于强化学习的锚泊辅助动力定位系统智能决策研究     

余尚禹  王磊  李博  衣凡 《海洋工程》2019,37(6):49-61

针对半潜平台锚泊辅助动力定位系统的最优定位点问题,设计了基于强化学习中深度神经网络的Q学习(DQN)控制策略的锚泊辅助动力定位的智能决策系统。该决策系统中DQN方法与比例—积分—微分(PID)控制方法相结合使用,实现系统优化。在基于机器人操作系统(ROS)平台的动力定位时域模拟程序中进行数值仿真,仿真结果验证了该系统在定位点决策问题上的可靠性和有效性,从而使半潜平台在面对未知海况时,均能寻找到最优定位点,在保证锚泊辅助动力定位系统可靠性的同时降低功率消耗,提高经济性。  相似文献   

Seismic fragilities of single-column highway bridges with rocking column-footing     

Yazhou Xie  Jian Zhang  Reginald DesRoches  Jamie E. Padgett 《地震工程与结构动力学》2019,48(7):843-864

Rocking isolation has been increasingly studied as a promising design concept to limit the earthquake damage of civil structures. Despite the difficulties and uncertainties of predicting the rocking response under individual earthquake excitations (due to negative rotational stiffness and complex impact energy loss), in a statistical sense, the seismic performance of rocking structures has been shown to be generally consistent with the experimental outcomes. To this end, this study assesses, in a probabilistic manner, the effectiveness of using rocking isolation as a retrofit strategy for single-column concrete box-girder highway bridges in California. Under earthquake excitation, the rocking bridge could experience multi-class responses (eg, full contacted or uplifting foundation) and multi-mode damage (eg, overturning, uplift impact, and column nonlinearity). A multi-step machine learning framework is developed to estimate the damage probability associated with each damage scenario. The framework consists of the dimensionally consistent generalized linear model for regression of seismic demand, the logistic regression for classification of distinct response classes, and the stepwise regression for feature selection of significant ground motion and structural parameters. Fragility curves are derived to predict the response class probabilities of rocking uplift and overturning, and the conditional damage probabilities such as column vibrational damage and rocking uplift impact damage. The fragility estimates of rocking bridges are compared with those for as-built bridges, indicating that rocking isolation is capable of reducing column damage potential. Additionally, there exists an optimal slenderness angle range that enables the studied bridges to experience much lower overturning tendencies and significantly reduced column damage probabilities at the same time.  相似文献   

基于长短时记忆神经网络的台风路径临近预报模型   总被引：3，自引：0，他引：3  

高松  赵鹏  潘斌  李亚汝  周敏  徐江玲  钟山  史振威 《海洋学报(英文版)》2018,37(5):8-12

It is of vital importance to reduce injuries and economic losses by accurate forecasts of typhoon tracks. A huge amount of typhoon observations have been accumulated by the meteorological department, however, they are yet to be adequately utilized. It is an effective method to employ machine learning to perform forecasts. A long short term memory(LSTM) neural network is trained based on the typhoon observations during 1949–2011 in China's Mainland, combined with big data and data mining technologies, and a forecast model based on machine learning for the prediction of typhoon tracks is developed. The results show that the employed algorithm produces desirable 6–24 h nowcasting of typhoon tracks with an improved precision.  相似文献   

基于Google Earth Engine与机器学习的省级尺度零散分布草地生物量估算   总被引：4，自引：0，他引：4  

修晓敏  周淑芳  陈黔  蒙继华  董文全  杨广斌  李晓松 《测绘通报》2019,(3):46-52,75

大区域草地地上生物量估算对草地资源利用管理及全球碳循环研究具有重要意义。为高效快速地估算大区域零散分布草地地上生物量，本文选取安徽省为研究区，在谷歌地球云引擎（Google Earth Engine）平台的支撑下，通过机器学习方法建立Landsat 8 OLI及其他辅助数据与地面实测草地地上生物量之间的联系，开展了草地零散分布地区省级尺度草地地上生物量高分辨率估算，并与传统的基于归一化植被指数（NDVI）回归模型进行了比较。研究结果表明，综合利用光谱与地形因子的机器学习方法，估算零散化分布草地地上生物量的精度可以达到65%以上，其中分类回归树（CART）模型R²=0.57，预测精度为68.60%，支持向量机（SVM）模型R²=0.59，预测精度为75.74%，而使用NDVI的回归分析产生的误差较大，R²=0.37，预测精度为57.51%，因此机器学习方法相对于传统基于NDVI的回归分析具有明显优势。另外，谷歌地球云引擎平台数据来源广泛、获取方便，可以高效地实现海量影像数据的预处理及计算分析，大大提升了工作效率，与地面调查数据的结合可实现更大区域乃至全国尺度上的零散分布草地地上生物量高分辨率遥感估算。  相似文献   

Deep crustal source of gneiss dome revealed by eclogite in migmatite (Montagne Noire,French Massif Central)     

Donna L. Whitney  Clémentine Hamelin  Christian Teyssier  Natalie H. Raia  Megan S. Korchinski  Nicholas C. A. Seaton  Brian C. Bagley  Anette von der Handt  Françoise Roger  Patrice F. Rey 《Journal of Metamorphic Geology》2020,38(3):297-327

In orogens worldwide and throughout geologic time, large volumes of deep continental crust have been exhumed in domal structures. Extension-driven ascent of bodies of deep, hot crust is a very efficient mechanism for rapid heat and mass transfer from deep to shallow crustal levels and is therefore an important mechanism in the evolution of continents. The dominant rock type in exhumed domes is quartzofeldspathic gneiss (typically migmatitic) that does not record its former high-pressure (HP) conditions in its equilibrium mineral assemblage; rather, it records the conditions of emplacement and cooling in the mid/shallow crust. Mafic rocks included in gneiss may, however, contain a fragmentary record of a HP history, and are evidence that their host rocks were also deeply sourced. An excellent example of exhumed deep crust that retains a partial HP record is in the Montagne Noire dome, French Massif Central, which contains well-preserved eclogite (garnet+omphacite+rutile+quartz) in migmatite in two locations: one in the dome core and the other at the dome margin. Both eclogites record P ~ 1.5 ± 0.2 GPa at T  ~  700 ± 20°C, but differ from each other in whole-rock and mineral composition, deformation features (shape and crystallographic preferred orientation, CPO), extent of record of prograde metamorphism in garnet and zircon, and degree of preservation of inherited zircon. Rim ages of zircon in both eclogites overlap with the oldest crystallization ages of host gneiss at c. 310 Ma, interpreted based on zircon rare earth element abundance in eclogite zircon as the age of HP metamorphism. Dome-margin eclogite zircon retains a widespread record of protolith age (c. 470–450 Ma, the same as host gneiss protolith age), whereas dome-core eclogite zircon has more scarce preservation of inherited zircon. Possible explanations for differences in the two eclogites relate to differences in the protolith mafic magma composition and history and/or the duration of metamorphic heating and extent of interaction with aqueous fluid, affecting zircon crystallization. Differences in HP deformation fabrics may relate to the position of the eclogite facies rocks relative to zones of transpression and transtension at an early stage of dome development. Regardless of differences, both eclogites experienced HP metamorphism and deformation in the deep crust at c. 310 Ma and were exhumed by lithospheric extension—with their host migmatite—near the end of the Variscan orogeny. The deep crust in this region was rapidly exhumed from ~50 to <10 km, where it equilibrated under low-P/high-T conditions, leaving a sparse but compelling record of the deep origin of most of the crust now exposed in the dome.  相似文献   

深海采矿循环式水力集矿系统设计与数值研究     

赵伟杰  肖龙飞  赵国成  刘明月 《海洋工程》2020,38(5):99-106

为使深海采矿作业更为高效、环保，提出了一种用于深海采矿的循环式水力集矿系统设计方案。该系统主要由集矿头、矿-沙分离器、水-沙分离器、水泵以及连接各部件的管道组成。采用基于欧拉法的液固两相流计算模型对该系统的关键问题，即矿-沙分离器和水-沙分离器的泥沙颗粒流动进行数值研究。相间拖曳力采用Richardson-Zaki模型。液相与固相湍流分别采用标准k-ε模型和Tchen湍流响应模型。该模型经验证对于直径0.165 mm与0.27 mm细颗粒具有较高精度。利用该模型，以0.165 mm泥沙颗粒为例，得到了两种分离器的分离效果与入口流速之间的规律，分析了两者颗粒浓度分布与流场特征。结果表明，两种分离器的性能均与入口流速有关，在选取适当入口条件时均可取得良好效果。  相似文献   

长江中下游成矿带铁-铜成矿系统结构的地球物理探测: 综合分析   总被引：1，自引：0，他引：1  

吕庆田  孟贵祥  严加永  张昆  龚雪婧  高凤霞 《地学前缘》2020,27(2):232-253

成矿系统是在深部过程驱动下形成的、具有自组织的能量及物质迁移-汇聚系统。系统在形成和演化过程中,在岩石圈不同尺度上留下“痕迹”,这种“痕迹”可以通过地球物理、地球化学和遥感等方法进行探测或观测。文章尝试在成矿系统理论框架下,对近10多年来在长江中下游成矿带进行的多尺度地球物理、地球化学探测结果进行分析,识别典型陆内成矿系统“源区”“通道”“场所”的地球物理、地球化学“痕迹”,尝试构建陆内成矿系统的空间结构模型。主要结论有:(1)长江中下游晚中生代的大规模铁、铜多金属成矿作用是一个完整的成矿系统。该系统包括3个子系统,分别为与高钾钙碱性岩浆岩有关的夕卡岩-斑岩成矿子系统、与橄榄安粗岩有关的陆相火山岩铁(硫)成矿子系统和与碱性岩有关的铜-金(铀)成矿子系统。(2)成矿系统的“源区”来自富集地幔的熔融、底侵,并在壳/幔边界与下地壳物质的混合,具有多级分布特点。幔源岩浆与地壳物质混合的比例决定了成矿金属的类型。(3)成矿带发育的“鳄鱼嘴”构造是铁铜成矿系统的主干“通道”。成矿系统“末端”矿质沉淀的“场所”主要受近地表褶皱、断裂、层间滑脱断层,以及由它们形成的断裂(裂隙)网络控制。(4)区域磁异常、放射性和地球化学异常是成矿系统残留“痕迹”的响应和“标识”。通过分析不同尺度的“标识”特征,可以深入认识成矿系统的空间结构,并可用于深部成矿预测。  相似文献   

淮南煤田刘庄煤矿深部煤层气地质特征     

窦新钊  章云根  张文永  孙贵  易小会  丁海 《中国煤炭地质》2020,(1):48-51

为了查明刘庄煤矿深部煤层气赋存及开发地质条件,在井田内实施了两口煤层气探井,并开展了系统的分析测试工作。结果表明:刘庄深部勘查区主要煤层孔隙度一般为4.14%~4.77%,比表面积集中在2.184~5.228m 2/g,13-1煤层、11-2煤层和8煤层试井渗透率分别为0.12mD、0.09mD和0.08mD,孔裂隙系统发育一般,属于渗透性差的储集层;储层压力梯度大于静水压力梯度,属高异常压力范畴;主要煤层兰氏压力平均为2.61MPa,兰氏体积平均为6.74m 3/t,吸附能力较低;LT-1井气测录井过程中共出现14层气测异常,异常层段全烃含量均较低,最大为3.701%(16-1煤);主要煤层的含气量分布在0.21~1.47m 3/t,平均0.65m 3/t,主要煤层含气饱和度均很低,最大值仅为18.0%。综合分析认为,刘庄煤矿深部主要煤层埋深大,孔裂隙系统发育差,渗透率低,而且具有低含气量和低饱和度的特征,煤层气勘探风险较高。  相似文献   

渤海湾盆地中南部干热岩选区方向     

张英  冯建赟  罗军  何治亮  武晓玲 《地学前缘》2020,27(1):35-47

渤海湾盆地是我国华北油气开发的主战场,多年的油气勘探开发积累了大量钻井、测录井及测温数据。通过对区内深钻井测温数据的分析研究,发现中南部的冀中坳陷、黄骅坳陷、临清坳陷和济阳坳陷是干热岩地热资源的主要富集区,温度≥180 ℃的热储主要发育在基岩潜山或斜坡区,埋深适宜,温度较高,有一定的天然孔渗条件。4个主要坳陷区4 000 m以深钻井平均地温梯度范围为3.07~3.32 ℃/100 m,5 000 m以深钻井平均地温梯度范围为2.96~3.27 ℃/100 m,埋深越大,地温梯度越低。坳陷区埋深在4 309~6 261 m时相继达到180 ℃温度,为干热岩目标埋深范围。不同坳陷、区块和井区达到180 ℃干热岩温度界限的深度相差较大。下古生界、中元古界和太古宇三套热储是渤海湾盆地中南部最具开发利用潜力的干热岩热储类型。冀中坳陷主要发育奥陶系灰岩、寒武系白云岩和中元古界白云岩热储,黄骅坳陷以奥陶系灰岩热储为主,济阳坳陷发育寒武系-奥陶系灰岩和太古宇变质岩热储,临清坳陷主要为奥陶系灰岩热储。坳陷区的凹中和凹边凸起区,顶面埋深3 000~5 000 m的前中生界潜山是主要的勘探方向,冀中坳陷河西务、长洋淀等潜山,黄骅坳陷南大港、千米桥等潜山,济阳坳陷桩西孤北等潜山,临清坳陷马场、文留等潜山是干热岩勘探开发有利区。干热岩勘探可与深层油气勘探开发相结合,进行不同类型资源的兼探和综合开发利用。  相似文献