基于Stacking集成学习的恒星/星系分类研究          下载免费PDF全文

李超  张文辉  李然  王俊义  林基明 《天文学报》2020,61(2):21-111

机器学习在当今诸多领域已经取得了巨大的成功,但是机器学习的预测效果往往依赖于具体问题.集成学习通过综合多个基分类器来预测结果,因此,其适应各种场景的能力较强,分类准确率较高.基于斯隆数字巡天(Sloan Digital Sky Survey,SDSS)计划恒星/星系中最暗源星等集分类正确率低的问题,提出一种基于Stacking集成学习的恒星/星系分类算法.从SDSS-DR7(SDSS Data Release 7)中获取完整的测光数据集,并根据星等值划分为亮源星等集、暗源星等集和最暗源星等集.仅针对分类较为复杂且困难的最暗源星等集展开分类研究.首先,对最暗源星等集使用10折嵌套交叉验证,然后使用支持向量机(Support Vector Machine,SVM)、随机森林(Random Forest,RF)、XGBoost(eXtreme Gradient Boosting)等算法建立基分类器模型;使用梯度提升树(Gradient Boosting Decision Tree,GBDT)作为元分类器模型.最后,使用基于星系的分类正确率等指标,与功能树(Function Tree,FT)、SVM、RF、GBDT、XGBoost、堆叠降噪自编码(Stacked Denoising AutoEncoders,SDAE)、深度置信网络(Deep Belief Network,DBN)、深度感知决策树(Deep Perception Decision Tree,DPDT)等模型进行分类结果对比分析.实验结果表明,Stacking集成学习模型在最暗源星等集分类中要比FT算法的星系分类正确率提高了将近10%.同其他传统的机器学习算法、较强的提升算法、深度学习算法相比,Stacking集成学习模型也有较大的提升.  相似文献   

Relation of Velocity Distribution and Mass for DA White Dwarfstwo     

Xiao-bo Gong  Li-fang Li 《Chinese Astronomy and Astrophysics》2018,42(2):202-212

White dwarfs are the evolutionary endpoint of the low-and-medium mass stars. In the studies of white dwarfs, the mass of white dwarf is an important physical parameter. In this paper, we give an analysis about the velocity distribution of DA white dwarfs in the Sloan Digital Sky Survey (SDSS), and hope to find the relation between mass and velocity distribution of white dwarfs. We get the radial velocity and tangential velocity of every DA white dwarf according to their proper motion and spectral shift. Through analyzing the velocity distribution of DA white dwarfs, we find that the small-mass white dwarfs, which are produced from the single-star evolution channel, have a relatively large velocity dispersion.  相似文献   

融合时间序列环境卫星数据与物候特征的水稻种植区提取   总被引：3，自引：0，他引：3  

柳文杰  曾永年  张猛 《遥感学报》2018,22(3):381-391

获取高精度的区域水稻种植面积对于农业规划、配置与决策具有重要意义。区域尺度的水稻面积获取依赖于高时空分辨率影像,但受卫星回访周期和气候影响,难以获取足够时间序列的高时空分辨率影像,从而影响水稻种植面积遥感提取的精度。为此,提出适应于中国南方多雨云天气地区,基于国产环境卫星(HJ-1A/1B)与MODIS融合数据的水稻种植面积提取的新方法。以洞庭湖区为实验区,利用STARFM模型融合环境卫星NDVI数据与MODIS13Q1数据,获取时间序列的环境卫星NDVI数据,利用水稻关键期的NDVI数据结合物候特征参数对水稻种植区域进行提取。结果表明,该方法能有效提取区域水稻种植的面积,水稻种植面积提取的总体精度与Kappa系数分别达到91.71%与0.9024,分类结果明显优于仅采用多光谱影像或NDVI数据。该研究为中国南方多雨云天气地区水稻种植面积提取提供了有效的方法。  相似文献   

湖泊与陆地雷电参数分布特征及其对比分析     

王学良  余田野  贺姗 《气象与环境学报》2019,35(6):132-138

根据湖北省雷电定位系统（Lightning Location System,LLS）2007年1月1日至2016年12月31日监测资料,采用数理统计方法,对湖泊和陆地区域的地闪频次、极性、地闪密度、雷电流幅值和波头陡度等雷电参数分布特征进行了对比研究。结果表明：湖泊与陆地的雷电参数时间变化趋势基本一致。近10 a闪电频次呈明显减少趋势,正地闪比例呈上升趋势;陆地比湖泊区域的闪电频次多,春夏季湖泊和陆地闪电频次差异明显,其中,夏季陆地闪电频次比湖泊多21.1%;湖泊和陆地闪电频次日变化大致呈单峰型,13-18时陆地闪电频次比湖泊多39.7%。湖泊地闪密度比陆地小,湖泊和陆地平均地闪密度分别为2.96次·（km^-2·a^-1）和3.47次·（km^-2·a^-1）。湖泊的平均雷电流幅值较陆地大;湖泊和陆地的平均雷电流波头陡度变化不大,相差一般在1 kA·（μs）^-1以下。  相似文献   

油气基础地质调查井新柯地1井工程设计          下载免费PDF全文

李正前  罗宏保  薛骁彤 《探矿工程》2018,45(8):39-43

以新柯地1井钻井工程设计为例,简述了工程设计的一般要求和原则,对新柯地1井工程设计的难点、钻井结构设计、钻机选型和设备要求、钻具组合和钻进参数、钻井液、固井、井控设计、录井、测井以及钻井工程施工情况进行了介绍。  相似文献   

FLAC3D在边坡数值分析及位移监测中应用     

张颖  孙兆涛  裴仲文 《探矿工程》2018,45(3):57-61

以辽宁省岫岩县某山头边坡为例,通过槽探、钻探、大型剪切试验等多种手段,找出合理潜在滑动面。利用FLAC3D软件对该边坡进行三维数值模拟,分析应力应变关系,得出变形较大及应力集中区域,同时利用强度折减法得出边坡的安全系数,评价其稳定性。将从边坡后缘张拉裂隙上布设的位移监测点采集到的数据,与数值模拟结果进行对比分析,验证数值模拟的可靠性,为边坡抗滑设计提供依据。  相似文献   

A Comprehensive Observational Analysis for the Effects of Gas Cannons on Clouds and Precipitation     

吴彬  王丹丹  李艳芳  尹浩  杜爽  黎朋红 《热带气象学报(英文版)》2022,28(2):237-251

To analyze the effects of gas cannons on clouds and precipitation, multisource observational data, including those from National Centers for Environmental Prediction (NCEP) reanalysis, Hangzhou and Huzhou new-generation weather radars, laser disdrometer, ground-based automatic weather station, wind profiler radar, and Lin''an C-band dual polarization radar, were adopted in this study. Based on the variational dual-Doppler wind retrieval method and the polarimetric variables obtained by the dual-polarization radar, we analyzed the microphysical processes and the variations in the macro - and microphysical quantities in clouds from the perspective of the synoptic background before precipitation enhancement, the polarization echo characteristics before, during and after enhancement, and the evolution of the fine three-dimensional kinematic structure and the microphysical structure. The results show that the precipitation enhancement operation promoted the development of radar echoes and prolonged their duration, and both the horizontal and vertical wind speeds increased. The dual-polarization radar echo showed that the diameter of the precipitation particles increased, and the concentration of raindrops increased after precipitation enhancement. The raindrops were lifted to a height corresponding to 0 to -20 ℃ due to vertical updrafts. Based on the disdrometer data during precipitation enhancement, the concentration of small raindrops (lgNw) showed a significant increase, and the mass-weighted diameter Dm value decreased, indicating that the precipitation enhancement operation played a certain“lubricating”effect. After the precipitation enhancement, the concentration of raindrops did not change much compared with that during the enhancement process, while the Dm increased, corresponding to an increase in rain intensity. The results suggest the positive effect of gas cannons on precipitation enhancement.  相似文献   

Experiments of DSAEF_LTP Model with Two Improved Parameters for Accumulated Precipitation of Landfalling Tropical Cyclones over Southeast China     

秦思  贾莉  丁晨晨  任福民  约翰·L·麦克布莱  李国平 《热带气象学报(英文版)》2022,28(3):286-296

The Dynamical-Statistical-Analog Ensemble Forecast model for landfalling tropical cyclones (TCs) precipitation (DSAEF_LTP) utilises an operational numerical weather prediction (NWP) model for the forecast track, while the precipitation forecast is obtained by finding analog cyclones, and making a precipitation forecast from an ensemble of the analogs. This study addresses TCs that occurred from 2004 to 2019 in Southeast China with 47 TCs as training samples and 18 TCs for independent forecast experiments. Experiments use four model versions. The control experiment DSAEF_LTP_1 includes three factors including TC track, landfall season, and TC intensity to determine analogs. Versions DSAEF_LTP_2, DSAEF_LTP_3, and DSAEF_LTP_4 respectively integrate improved similarity region, improved ensemble method, and improvements in both parameters. Results show that the DSAEF_LTP model with new values of similarity region and ensemble method (DSAEF_LTP_4) performs best in the simulation experiment, while the DSAEF_LTP model with new values only of ensemble method (DSAEF_LTP_3) performs best in the forecast experiment. The reason for the difference between simulation (training sample) and forecast (independent sample) may be that the proportion of TC with typical tracks (southeast to northwest movement or landfall over Southeast China) has changed significantly between samples. Forecast performance is compared with that of three global dynamical models (ECMWF, GRAPES, and GFS) and a regional dynamical model (SMS-WARMS). The DSAEF_LTP model performs better than the dynamical models and tends to produce more false alarms in accumulated forecast precipitation above 250 mm and 100 mm. Compared with TCs without heavy precipitation or typical tracks, TCs with these characteristics are better forecasted by the DSAEF_LTP model.  相似文献   

Characteristics of Radar Echo Parameters and Microphysical Structure Simulation of a Short-Time Heavy Precipitation Supercell     

赵桂香  王一颉  连志鸾 《热带气象学报(英文版)》2022,28(4):388-404

By using the conventional observations, radar data, NCEP/NCAR FNL 1o×1o reanalysis data and numerical simulation data and with the construction and calculation of radar echo parameters, this paper presents the structural characteristics and physical processes of a short-time heavy precipitation supercell that occurred in the squall line process in Shanxi Province on 24 June 2020. The results show that this squall line event occurred in front of a surface cold front, combined with infiltration of low-level cold air and continuous increase of near-surface humidity in the afternoon. The surface mesoscale convergence line and mesoscale dew point front contributed to the development and systemization of the squall line by a large degree. The short-time extremely heavy precipitation in Pingshun County was caused by the development of a supercell from thunderstorm cells on the front side of the squall line. The characteristics of sharp increase in vertical integral liquid water content, persistent increase in reflectivity factor and continuous rise in the echo top height appeared about 23 min earlier than the severe precipitation, which has qualitative indicating significance for the nowcasting of short-time heavy precipitation. A quantitative analysis of the radar echo parameters suggests that the “sudden drop”of FV40 was a precursor signal of cells’coalescence and rapid development to the mature stage. The areal change of the echo core at the 6 km height was highly subject to the merging and developing of cells, the rapid change of hydrometeor particles in clouds and the precipitation intensity. Changes in the cross-sectional area of convective cells at different heights can indirectly reflect the changes of liquid particles and ice particles in clouds, which is indicatively meaningful for predicting the coalescing and developing-to-maturing of cells and heavy precipitation 30-45 min earlier. A comprehensive echo parameter prediction model constructed by the random forest principle can predict the magnitude of short-time heavy precipitation 40-50 min in advance. Numerical simulation reveals that large amounts of water vapor existed in the near-surface atmosphere, and that the cells rapidly obtained moisture from the ambient atmosphere and developed rapidly through maternal feeding. The cold cloud zone was narrow, upright and had a high stretch height. The upward motion in clouds was strong and deep, and very rich in liquid water content. The graupel particles had a large vertical distribution range, the coexistence area of graupel and snow was large, the height of raindrops was close to the surface with a wide horizontal scale, and the precipitation efficiency was high. These may be the important elements responsible for the occurrence of the short-time heavy precipitation that exceeded historical extreme values. On the basis of the above analyses, a comprehensive parameter (CP) prediction model is worked out, which can estimate the developing trend of supercells and the intensity of short-time heavy precipitation about 1 h in advance.  相似文献   

2020年7月12日唐山5.1级地震前高光谱多参数异常特征分析          下载免费PDF全文

丁志华  崔月菊  唐杰 《中国地震》2022,38(3):494-502

基于高光谱遥感数据,通过RST算法提取2020年7月12日唐山5.1级地震前不同参数异常信息,发现震前CH₄、CO、H₂O、O₃均出现了高值异常,且异常均位于震中附近,异常幅度高于2倍均方差,CH₄、CO、H₂O异常机理不同于O₃。通过统计2018—2020年CH₄、CO、H₂O三种气体出现频次,发现同步异常具有可靠性,认为高光谱多参数异常出现的同步性,可提高异常性质判定的信度。通过对不同参数进行连续跟踪观测,提取多参数同步异常信息,对地震短临预报具有一定参考意义。  相似文献