深度学习是当前机器学习、模式识别和人工智能领域中的一项热点研究技术,非常适用于处理复杂的大规模数据.基于深度学习理论构建了一个5层的栈式自编码深度神经网络,对恒星大气物理参数进行自动估计,网络各层的节点数分别为3821-500-100...深度学习是当前机器学习、模式识别和人工智能领域中的一项热点研究技术,非常适用于处理复杂的大规模数据.基于深度学习理论构建了一个5层的栈式自编码深度神经网络,对恒星大气物理参数进行自动估计,网络各层的节点数分别为3821-500-100-50-1.使用美国大型巡天项目Sloan发布的Sloan Digital Sky Survey(SDSS)实测光谱以及由Kurucz的New Opacity Distribution Function(NEWODF)模型得到的理论光谱进行了实验验证,对有效温度(Teff)、表面重力加速度(lg g)和金属丰度([Fe/H])3个物理参数进行了自动估计.结果表明,栈式自编码深度神经网络的估计精度较好,其中在SDSS数据上的平均绝对误差分别为:79.95(Teff/K),0.0058(lg(Teff/K)),0.1706(lg(g/(cm·s^(-2)))),0.1294 dex([Fe/H]);在理论数据上的平均绝对误差分别是:15.34(Teff/K),0.0011(lg(Teff/K)),0.0214(lg(g/(cm·s^(-2)))),0.0121 dex([Fe/H]).展开更多
We report on the first investigation into kinematics and chromospheric activity of M dwarfs from the Guo Shou Jing Telescope （also called the Large Sky Area Multi-Object Fiber Spectroscopic Telescope - LAMOST） data ...We report on the first investigation into kinematics and chromospheric activity of M dwarfs from the Guo Shou Jing Telescope （also called the Large Sky Area Multi-Object Fiber Spectroscopic Telescope - LAMOST） data release one （DR1）. The sample comprises 71 304 M dwarfs. Their fundamental parameters such as spectral types, radial velocities, important molecular band indices and magnetic activities are measured. Their distances are determined by a spectroscopic parallax relation. Space motion （U, V, W） and Galactocentric cylindrical coordinates （R, θ, Z） for the M dwarfs are also computed. We examine velocity dispersion as a function of height from the Galactic plane and find that all three components of velocity dispersion in- crease with height as measured with respect to the Galactic plane. The investigation into chromospheric activities along the height from the Galactic plane confirms that M dwarfs closer to the Galactic plane are more likely to be active. We take a pure kinematical approach to select thin disk stars and thick disk stars from our sample, then to investigate the differences in properties between these two populations. Our analysis is in excellent agreement with previous studies and leads to a better understanding of the structure of the Galactic disk.展开更多
近年来,深度学习算法以其适应性强、准确率高、结构复杂等特性在数据挖掘算法中异军突起,但是在天文信息学中深度学习算法还鲜有问津.针对斯隆数字巡天(Sloan Digital Sky Survey,SDSS)恒星/星系分类中普遍存在的亮源集分类正确率高但...近年来,深度学习算法以其适应性强、准确率高、结构复杂等特性在数据挖掘算法中异军突起,但是在天文信息学中深度学习算法还鲜有问津.针对斯隆数字巡天(Sloan Digital Sky Survey,SDSS)恒星/星系分类中普遍存在的亮源集分类正确率高但暗源集分类正确率低等问题,引入了深度学习中较新的研究成果—堆叠降噪自编码(stacked denoising autoencoders,SDA)神经网络和dropout微调技术.从SDSS释放出的带有光谱证认(spectroscopic measurements)的测光数据中分别随机抽取DR7(Data Release7)和DR12(Data Release 12)的亮源集和暗源集并对其进行预处理,再分别对它们的亮源集和暗源集做不放回随机抽样,得到它们亮源和暗源的训练集和测试集.最后用这些训练集分别训练得到了DR7和DR12亮源和暗源的SDA模型,并将SDA在DR12测试集上的测试结果与支持向量机软件包(Library for Support Vector Machines,LibSVM)、J48决策树(J48)、逻辑模型树(Logistic Model Trees,LMT)、支持向量机(Support Vector Machine,SVM)、逻辑回归(Logistic Regression)、单层决策树算法(Decision Stump)上的测试结果进行比较,同时将SDA在DR7测试集上的测试结果与6种决策树的测试结果进行比较.仿真表明SDA在SDSS-DR7和最新SDSS-DR12的暗源集上的分类性能明显优于其他算法,尤其是在使用完备函数(completeness function,CP)作为衡量指标时,SDA相比决策树算法在SDSS-DR7暗源集正确率提高了15%左右.展开更多
The statistical parameters of five generalizations of the Lindley distribution, such as the average, variance and moments, are reviewed. A new double truncated Lindley distribution with three parameters is derived. Th...The statistical parameters of five generalizations of the Lindley distribution, such as the average, variance and moments, are reviewed. A new double truncated Lindley distribution with three parameters is derived. The new distributions are applied to model the initial mass function for stars.展开更多
恒星光谱分类是天文学中一个重要的研究问题.对于已经采集到的海量高维恒星光谱数据的分类,采用模式匹配方法对光谱型分类较为成功,但其缺点在于标准恒星模版之间的差异性在匹配实际观测数据中不能体现出来,尤其是当需要进行光谱型和光...恒星光谱分类是天文学中一个重要的研究问题.对于已经采集到的海量高维恒星光谱数据的分类,采用模式匹配方法对光谱型分类较为成功,但其缺点在于标准恒星模版之间的差异性在匹配实际观测数据中不能体现出来,尤其是当需要进行光谱型和光度型的二元分类时模版匹配法往往会失败.而采用谱线特征测量的光度型分类强烈地依赖谱线拟合的准确性.为了解决二元分类的问题,介绍了一种基于卷积神经网络的恒星光谱型和光度型分类模型(Classification model of Stellar Spectral type and Luminosity type based on Convolution Neural Network,CSSL CNN).这一模型使用卷积神经网络来提取光谱的特征,通过注意力模块学习到了重要的光谱特征,借助池化操作降低了光谱的维度并压缩了模型参数的数量,使用全连接层来学习特征并对恒星光谱进行分类.实验中使用了大天区面积多目标光纤光谱天文望远镜(Large Sky Area Multi-Object Fiber Spectroscopy Telescope,LAMOST)公开数据集Data Release 5(DR5,用了其中71282条恒星光谱数据,每条光谱包含了3000多维的特征)对该模型的性能进行验证与评估.实验结果表明,基于卷积神经网络的模型在恒星的光谱型分类上准确率达到92.04%,而基于深度神经网络的模型(Celestial bodies Spectral Classification Model,CSC Model)只有87.54%的准确率;CSSL CNN在恒星的光谱型和光度型二元分类上准确率达到83.91%,而模式匹配方法MKCLASS仅有38.38%的准确率且效率较低.展开更多
We present LAMOST data on 168 γ Doradus(γ Dor) pulsating stars including stellar atmospheric parameters of 137 variables and spectral types for all of the samples. The distributions of period(P), temperature(T), gra...We present LAMOST data on 168 γ Doradus(γ Dor) pulsating stars including stellar atmospheric parameters of 137 variables and spectral types for all of the samples. The distributions of period(P), temperature(T), gravitational acceleration(log(g)) and metallicity [Fe/H] are shown. It is found that most γ Dor variables are main-sequence stars with early F spectral types and temperatures from 6880 K to7280 K. They are slightly more metal poor than the Sun with a metallicity range from-0.4 to 0. On the H-R and log g-T diagrams, both the γ Dor and δ Scuti(δ Sct) stars occupy in the same region and some are beyond the borders predicted by current stellar pulsation theories. It is discovered that the physical properties of γ Dor stars are similar to those of long-period δ Sct(P > 0.3 d) stars. The stellar atmospheric parameters are all correlated with the pulsation period for short-period δ Sct variables(P < 0.3 d), but there are no such relations for γ Dor or long-period δ Sct stars. These results reveal that γ Dor and long-period δ Sct are the same group of pulsating stars and they are different from short-period δ Sct variables. Meanwhile, 33γ Dor stars are identified as candidates of binary or multiple systems.展开更多
We present the complete set of physical and geometrical parameters of the visual close binary system GJ 9830 for the first time by applying Al-Wardat’s complex method.This method combines magnitude difference from sp...We present the complete set of physical and geometrical parameters of the visual close binary system GJ 9830 for the first time by applying Al-Wardat’s complex method.This method combines magnitude difference from speckle interferometry,synthetic spectral energy distributions of the binary components which are constructed based on grids of Kurucz blanketed models(ATLAS9),and the orbital solution using Tokovinin’s dynamical method to estimate the parameters of individual components.The analysis of the system by employing synthetic photometry resulted in the following set of parameters:Teff= 6220 ± 100 K,log g = 4.30 ± 0.12,R = 1.10 ± 0.08 R⊙for the primary component and Teff= 4870 ± 100 K,log g = 4.60 ± 0.11,R = 0.709 ± 0.07 R⊙for the secondary component.The recently published dynamical parallax from the Gaia space mission was used to calculate the total mass of the binary system as 1.75 ± 0.06 M⊙,which coincides with those estimated using Al-Wardat’s method as MA= 1.18 ± 0.10 M⊙,MB= 0.75 ± 0.08 M⊙.The analysis of the system reveals that both components are characteristic of main sequence stars and have an age of around 1.4 ± 0.50 Gyr.The evolutionary tracks and isochrones of the system’s components are discussed,and the fragmentation process is suggested as the most likely process for the formation of the system.展开更多
天体光谱分类是天文学研究的重要内容之一,其关键是从光谱数据中选择和提取对分类识别最有效的特征构建特征空间.提出一种新的基于2维傅里叶谱图像的特征提取方法,并应用于LAMOST (the Large Sky Area Multi-Object Fiber Spectroscopic...天体光谱分类是天文学研究的重要内容之一,其关键是从光谱数据中选择和提取对分类识别最有效的特征构建特征空间.提出一种新的基于2维傅里叶谱图像的特征提取方法,并应用于LAMOST (the Large Sky Area Multi-Object Fiber Spectroscopic Telescope)恒星光谱数据的分类研究中.光谱数据来源于LAMOST Data Release 5(DR5),选取30000条F、 G和K型星光谱数据,利用短时傅里叶变换(Short-Time Fourier Transform, STFT)将1维光谱数据变换成2维傅里叶谱图像,对得到的2维傅里叶谱图像采用深度卷积网络模型进行分类,得到的分类准确率是92.90%.实验结果表明通过对LAMOST恒星光谱数据进行STFT可得到光谱的2维傅里叶谱图像,谱图像构成了新的光谱数据特征和特征空间,新的特征对于光谱数据分类是有效的.此方法是对光谱分类的一种全新尝试,对海量天体光谱的分类和挖掘处理有一定的开创意义.展开更多
The aim of this paper is to determinate the fundamental parameters of six exoplanet host （EH） stars and their planets. Because techniques for detecting exo- planets yield properties of the planet only as a function ...The aim of this paper is to determinate the fundamental parameters of six exoplanet host （EH） stars and their planets. Because techniques for detecting exo- planets yield properties of the planet only as a function of the properties of the host star, we must accurately determine the parameters of the EH stars first. For this rea- son, we constructed a grid of stellar models including diffusion and rotation-induced extra-mixing with given ranges of input parameters （i.e. mass, metallicity and initial rotation rate）. In addition to the commonly used observational constraints such as the effective temperature Tell, luminosity L and metallicity [Fe/H], we added two obser- vational constraints, the lithium abundance log N （Li） and the rotational period Prot. These two additional observed parameters can set further constraints on the model due to their correlations with mass, age and other stellar properties. Hence, our estimations of the fundamental parameters for these EH stars and their planets have a higher preci- sion than previous works. Therefore, the combination of rotational period and lithium helps us to obtain more accurate parameters for stars, leading to an improvement in knowledge about the physical state of EH stars and their planets.展开更多
About 786.4 thousand stars were observed by LAMOST twice or more during the first stage of its spectroscopic survey. The radial velocity differences for about 256 thousand targets are larger than10 km s^(-1) and they ...About 786.4 thousand stars were observed by LAMOST twice or more during the first stage of its spectroscopic survey. The radial velocity differences for about 256 thousand targets are larger than10 km s^(-1) and they are possible spectroscopic binary or variable candidates(SBVCs). It is shown that most SBVCs are slightly metal poorer than the Sun. There are two peaks in the temperature distribution of SBVCs around 5760 K and 4870 K, while there are three peaks in the distribution of the gravitational acceleration at 2.461, 4.171 and 4.621 cm s^(-2). The locations of SBVCs on the [Fe/H]-T, [Fe/H]-log g, log g-T and H-R diagrams are investigated. It is found that the detected SBVCs could be classified into four groups. The first group has higher log g～4.621 and lower T ～ 4870 K which are mainly cool red dwarf binaries. The second group of SBVCs has logg around 4.171 cm s^(-2) that includes binaries and pulsating stars such as δSet and γ Dor variables. The gravitational accelerations of the third group of SBVCs are higher and some of them are below the zero-age main sequence. They may be contact binaries in which the primary components are losing energy to the secondaries in the common envelopes and are at a special stellar evolutionary stage.The last group is composed of giants or supergiants with log g around 2.461 cm s^(-2) that may be evolved pulsating stars. One target(C134624.29+333921.2) is confirmed as an eclipsing binary with a period of 0.65 days. A preliminary analysis suggests that it is a detached binary with a mass ratio of 0.46. The primary fills its critical Roche lobe by about 89%, indicating that mass transfer will occur between the two components.展开更多
We present the physical parameters of three short period close binaries using data observed from the Kepler Space Telescope. All of these observations were taken in a single bandpass(which approximates the Johnson V-b...We present the physical parameters of three short period close binaries using data observed from the Kepler Space Telescope. All of these observations were taken in a single bandpass(which approximates the Johnson V-band). Our three systems are KIC 2715417, KIC 6050116 and KIC 6287172. The first system, KIC 2715417, is considered a semi-detached system with the secondary component filling its Roche lobe. The second system, KIC 6050116, is an overcontact system, while the third system, KIC 6287172, belongs to ellipsoidal variables as deduced from the Roche lobe geometry. For photometric analysis, we used the PHOEBE software package, which is based on the Wilson-Devinney code. Due to lack of spectroscopic data, the photometric mass ratios are determined from the analyses of light curves using the q-search method. The absolute parameters are determined using three different methods(Harmanec, Maceroni &Van'tVeer and Gazeas & Niarchos).展开更多
We present photometric observations in Sloan filters g′, i′of the eclipsing W UMa stars USNOA2.0 1350-17365531, V471 Cas, V479 Lac and V560 Lac. The sinusoidal-like O-C diagram of V471 Cas indicates the presence of ...We present photometric observations in Sloan filters g′, i′of the eclipsing W UMa stars USNOA2.0 1350-17365531, V471 Cas, V479 Lac and V560 Lac. The sinusoidal-like O-C diagram of V471 Cas indicates the presence of a third body with mass 0.12 M_⊙(a red dwarf) at distance 897 R_⊙. The O-C diagram of V479 Lac reveals a period decrease of d P/dt =-1.69 × 10-6d yr-1. The results of the light curve solutions are:(i) the targets are overcontact binaries with small fill-out factors;(ii) their components are F–K stars, comparable in size, whose temperature differences are below 80 K;(iii) all targets undergo partial eclipses and to limit the possible mass ratios we carried out two-step q-search analysis. The target global parameters(luminosities, radii, masses) were obtained on the basis of their Gaia distances and the results of our light curve solutions. The obtained total mass of V560 Lac turns out to be smaller than the lower mass limit for presently known W UMa binaries of 1.0-1.2 M_⊙, i.e. this target is a peculiar overcontact system.展开更多
The best physical and geometrical parameters of the main sequence close visual binary system(CVBS), HIP 105947, are presented. These parameters have been constructed conclusively using Al-Wardat’s complex method for ...The best physical and geometrical parameters of the main sequence close visual binary system(CVBS), HIP 105947, are presented. These parameters have been constructed conclusively using Al-Wardat’s complex method for analyzing CVBSs, which is a method for constructing a synthetic spectral energy distribution(SED) for the entire binary system using individual SEDs for each component star. The model atmospheres are in its turn built using the Kurucz(ATLAS9) line-blanketed plane-parallel models. At the same time, the orbital parameters for the system are calculated using Tokovinin’s dynamical method for constructing the best orbits of an interferometric binary system. Moreover, the mass-sum of the components, as well as the ?θ and ?ρ residuals for the system, is introduced. The combination of Al-Wardat’s and Tokovinin’s methods yields the best estimations of the physical and geometrical parameters. The positions of the components in the system on the evolutionary tracks and isochrones are plotted and the formation and evolution of the system are discussed.展开更多
We present the physical and geometrical parameters of the individual components of the close visual double-lined spectroscopic binary system Gliese 762.1, which were estimated using Al-Wardat＇s complex method for ana...We present the physical and geometrical parameters of the individual components of the close visual double-lined spectroscopic binary system Gliese 762.1, which were estimated using Al-Wardat＇s complex method for analyzing close visual binary systems. The estimated parameters of the individual components of the system are as follows： radius RA= 0.845±0.09R⊙, RB= 0.795±0.10R⊙, effective temperature Teff^A = 5300luminosity±50 K, Teff^B= 5150 L±50 K, surface gravity log gA= 4.52±0.10, log gB=4.54±0.15 and A= 0.51 with a semi-major axis of 0.0865±0.08L⊙, LB= 0.40±0.07L⊙. New orbital elements are presented±0.010 arcsec using the Hippracos parallax π = 58.96 ndividual masses of the system are determined as M = 1.±0.65 mas, and an accurate total mass and i72±0.60 M⊙,MA= 0.89±0.08 M⊙and MB= 0.83 K0 V and±0.07 M⊙. Finally, the spectral types and luminosity classes of both components are assigned as K1.5V for the primary and secondary components respectively,and their positions on the H-R diagram and evolutionary tracks are given.展开更多
Using stellar evolutionary models, we investigate the effects of convective overshooting on naked helium stars. We find that a larger value of overshooting parameter δov results in a larger convective core, which pro...Using stellar evolutionary models, we investigate the effects of convective overshooting on naked helium stars. We find that a larger value of overshooting parameter δov results in a larger convective core, which prolongs the lifetimes of naked helium stars on the helium main sequence and leads to higher effective temperatures and luminosities. For naked helium stars with masses lower than about 0.8 Mo, they hardly become giant stars as a result of a weak burning shell. However, naked helium stars with masses between about 0.8 M⊙ and 1.1 M⊙ can evolve into giant branch phases, and finally become carbon oxygen white dwarfs.展开更多
文摘深度学习是当前机器学习、模式识别和人工智能领域中的一项热点研究技术,非常适用于处理复杂的大规模数据.基于深度学习理论构建了一个5层的栈式自编码深度神经网络,对恒星大气物理参数进行自动估计,网络各层的节点数分别为3821-500-100-50-1.使用美国大型巡天项目Sloan发布的Sloan Digital Sky Survey(SDSS)实测光谱以及由Kurucz的New Opacity Distribution Function(NEWODF)模型得到的理论光谱进行了实验验证,对有效温度(Teff)、表面重力加速度(lg g)和金属丰度([Fe/H])3个物理参数进行了自动估计.结果表明,栈式自编码深度神经网络的估计精度较好,其中在SDSS数据上的平均绝对误差分别为:79.95(Teff/K),0.0058(lg(Teff/K)),0.1706(lg(g/(cm·s^(-2)))),0.1294 dex([Fe/H]);在理论数据上的平均绝对误差分别是:15.34(Teff/K),0.0011(lg(Teff/K)),0.0214(lg(g/(cm·s^(-2)))),0.0121 dex([Fe/H]).
基金Supported by the National Natural Science Foundation of China
文摘We report on the first investigation into kinematics and chromospheric activity of M dwarfs from the Guo Shou Jing Telescope （also called the Large Sky Area Multi-Object Fiber Spectroscopic Telescope - LAMOST） data release one （DR1）. The sample comprises 71 304 M dwarfs. Their fundamental parameters such as spectral types, radial velocities, important molecular band indices and magnetic activities are measured. Their distances are determined by a spectroscopic parallax relation. Space motion （U, V, W） and Galactocentric cylindrical coordinates （R, θ, Z） for the M dwarfs are also computed. We examine velocity dispersion as a function of height from the Galactic plane and find that all three components of velocity dispersion in- crease with height as measured with respect to the Galactic plane. The investigation into chromospheric activities along the height from the Galactic plane confirms that M dwarfs closer to the Galactic plane are more likely to be active. We take a pure kinematical approach to select thin disk stars and thick disk stars from our sample, then to investigate the differences in properties between these two populations. Our analysis is in excellent agreement with previous studies and leads to a better understanding of the structure of the Galactic disk.
文摘The statistical parameters of five generalizations of the Lindley distribution, such as the average, variance and moments, are reviewed. A new double truncated Lindley distribution with three parameters is derived. The new distributions are applied to model the initial mass function for stars.
文摘恒星光谱分类是天文学中一个重要的研究问题.对于已经采集到的海量高维恒星光谱数据的分类,采用模式匹配方法对光谱型分类较为成功,但其缺点在于标准恒星模版之间的差异性在匹配实际观测数据中不能体现出来,尤其是当需要进行光谱型和光度型的二元分类时模版匹配法往往会失败.而采用谱线特征测量的光度型分类强烈地依赖谱线拟合的准确性.为了解决二元分类的问题,介绍了一种基于卷积神经网络的恒星光谱型和光度型分类模型(Classification model of Stellar Spectral type and Luminosity type based on Convolution Neural Network,CSSL CNN).这一模型使用卷积神经网络来提取光谱的特征,通过注意力模块学习到了重要的光谱特征,借助池化操作降低了光谱的维度并压缩了模型参数的数量,使用全连接层来学习特征并对恒星光谱进行分类.实验中使用了大天区面积多目标光纤光谱天文望远镜(Large Sky Area Multi-Object Fiber Spectroscopy Telescope,LAMOST)公开数据集Data Release 5(DR5,用了其中71282条恒星光谱数据,每条光谱包含了3000多维的特征)对该模型的性能进行验证与评估.实验结果表明,基于卷积神经网络的模型在恒星的光谱型分类上准确率达到92.04%,而基于深度神经网络的模型(Celestial bodies Spectral Classification Model,CSC Model)只有87.54%的准确率;CSSL CNN在恒星的光谱型和光度型二元分类上准确率达到83.91%,而模式匹配方法MKCLASS仅有38.38%的准确率且效率较低.
基金Funding for the project has been provided by the National Development and Reform CommissionFunding for the DPAC has been provided by national institutions,in particular the institutions participating in the Gaia Multilateral Agreement
文摘We present LAMOST data on 168 γ Doradus(γ Dor) pulsating stars including stellar atmospheric parameters of 137 variables and spectral types for all of the samples. The distributions of period(P), temperature(T), gravitational acceleration(log(g)) and metallicity [Fe/H] are shown. It is found that most γ Dor variables are main-sequence stars with early F spectral types and temperatures from 6880 K to7280 K. They are slightly more metal poor than the Sun with a metallicity range from-0.4 to 0. On the H-R and log g-T diagrams, both the γ Dor and δ Scuti(δ Sct) stars occupy in the same region and some are beyond the borders predicted by current stellar pulsation theories. It is discovered that the physical properties of γ Dor stars are similar to those of long-period δ Sct(P > 0.3 d) stars. The stellar atmospheric parameters are all correlated with the pulsation period for short-period δ Sct variables(P < 0.3 d), but there are no such relations for γ Dor or long-period δ Sct stars. These results reveal that γ Dor and long-period δ Sct are the same group of pulsating stars and they are different from short-period δ Sct variables. Meanwhile, 33γ Dor stars are identified as candidates of binary or multiple systems.
文摘We present the complete set of physical and geometrical parameters of the visual close binary system GJ 9830 for the first time by applying Al-Wardat’s complex method.This method combines magnitude difference from speckle interferometry,synthetic spectral energy distributions of the binary components which are constructed based on grids of Kurucz blanketed models(ATLAS9),and the orbital solution using Tokovinin’s dynamical method to estimate the parameters of individual components.The analysis of the system by employing synthetic photometry resulted in the following set of parameters:Teff= 6220 ± 100 K,log g = 4.30 ± 0.12,R = 1.10 ± 0.08 R⊙for the primary component and Teff= 4870 ± 100 K,log g = 4.60 ± 0.11,R = 0.709 ± 0.07 R⊙for the secondary component.The recently published dynamical parallax from the Gaia space mission was used to calculate the total mass of the binary system as 1.75 ± 0.06 M⊙,which coincides with those estimated using Al-Wardat’s method as MA= 1.18 ± 0.10 M⊙,MB= 0.75 ± 0.08 M⊙.The analysis of the system reveals that both components are characteristic of main sequence stars and have an age of around 1.4 ± 0.50 Gyr.The evolutionary tracks and isochrones of the system’s components are discussed,and the fragmentation process is suggested as the most likely process for the formation of the system.
文摘天体光谱分类是天文学研究的重要内容之一,其关键是从光谱数据中选择和提取对分类识别最有效的特征构建特征空间.提出一种新的基于2维傅里叶谱图像的特征提取方法,并应用于LAMOST (the Large Sky Area Multi-Object Fiber Spectroscopic Telescope)恒星光谱数据的分类研究中.光谱数据来源于LAMOST Data Release 5(DR5),选取30000条F、 G和K型星光谱数据,利用短时傅里叶变换(Short-Time Fourier Transform, STFT)将1维光谱数据变换成2维傅里叶谱图像,对得到的2维傅里叶谱图像采用深度卷积网络模型进行分类,得到的分类准确率是92.90%.实验结果表明通过对LAMOST恒星光谱数据进行STFT可得到光谱的2维傅里叶谱图像,谱图像构成了新的光谱数据特征和特征空间,新的特征对于光谱数据分类是有效的.此方法是对光谱分类的一种全新尝试,对海量天体光谱的分类和挖掘处理有一定的开创意义.
基金Supported by the National Natural Science Foundation of China
文摘The aim of this paper is to determinate the fundamental parameters of six exoplanet host （EH） stars and their planets. Because techniques for detecting exo- planets yield properties of the planet only as a function of the properties of the host star, we must accurately determine the parameters of the EH stars first. For this rea- son, we constructed a grid of stellar models including diffusion and rotation-induced extra-mixing with given ranges of input parameters （i.e. mass, metallicity and initial rotation rate）. In addition to the commonly used observational constraints such as the effective temperature Tell, luminosity L and metallicity [Fe/H], we added two obser- vational constraints, the lithium abundance log N （Li） and the rotational period Prot. These two additional observed parameters can set further constraints on the model due to their correlations with mass, age and other stellar properties. Hence, our estimations of the fundamental parameters for these EH stars and their planets have a higher preci- sion than previous works. Therefore, the combination of rotational period and lithium helps us to obtain more accurate parameters for stars, leading to an improvement in knowledge about the physical state of EH stars and their planets.
基金The Guo Shou Jing Telescope (the Large Sky Area Multi-Object Fiber Spectroscopic Telescope, LAMOST) is a National Major Scientific Project built by the Chinese Academy of Sciencesprovided by the National Development and Reform Commission
文摘About 786.4 thousand stars were observed by LAMOST twice or more during the first stage of its spectroscopic survey. The radial velocity differences for about 256 thousand targets are larger than10 km s^(-1) and they are possible spectroscopic binary or variable candidates(SBVCs). It is shown that most SBVCs are slightly metal poorer than the Sun. There are two peaks in the temperature distribution of SBVCs around 5760 K and 4870 K, while there are three peaks in the distribution of the gravitational acceleration at 2.461, 4.171 and 4.621 cm s^(-2). The locations of SBVCs on the [Fe/H]-T, [Fe/H]-log g, log g-T and H-R diagrams are investigated. It is found that the detected SBVCs could be classified into four groups. The first group has higher log g～4.621 and lower T ～ 4870 K which are mainly cool red dwarf binaries. The second group of SBVCs has logg around 4.171 cm s^(-2) that includes binaries and pulsating stars such as δSet and γ Dor variables. The gravitational accelerations of the third group of SBVCs are higher and some of them are below the zero-age main sequence. They may be contact binaries in which the primary components are losing energy to the secondaries in the common envelopes and are at a special stellar evolutionary stage.The last group is composed of giants or supergiants with log g around 2.461 cm s^(-2) that may be evolved pulsating stars. One target(C134624.29+333921.2) is confirmed as an eclipsing binary with a period of 0.65 days. A preliminary analysis suggests that it is a detached binary with a mass ratio of 0.46. The primary fills its critical Roche lobe by about 89%, indicating that mass transfer will occur between the two components.
文摘We present the physical parameters of three short period close binaries using data observed from the Kepler Space Telescope. All of these observations were taken in a single bandpass(which approximates the Johnson V-band). Our three systems are KIC 2715417, KIC 6050116 and KIC 6287172. The first system, KIC 2715417, is considered a semi-detached system with the secondary component filling its Roche lobe. The second system, KIC 6050116, is an overcontact system, while the third system, KIC 6287172, belongs to ellipsoidal variables as deduced from the Roche lobe geometry. For photometric analysis, we used the PHOEBE software package, which is based on the Wilson-Devinney code. Due to lack of spectroscopic data, the photometric mass ratios are determined from the analyses of light curves using the q-search method. The absolute parameters are determined using three different methods(Harmanec, Maceroni &Van'tVeer and Gazeas & Niarchos).
基金supported partly by project DN08/20 of the Scientific Foundation of the Bulgarian Ministry of Education and Scienceby project RD 08-142 of Shumen University+1 种基金the support of the private IRIDA Observatory operated remotely(www.iridaobservatory.org)Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement
文摘We present photometric observations in Sloan filters g′, i′of the eclipsing W UMa stars USNOA2.0 1350-17365531, V471 Cas, V479 Lac and V560 Lac. The sinusoidal-like O-C diagram of V471 Cas indicates the presence of a third body with mass 0.12 M_⊙(a red dwarf) at distance 897 R_⊙. The O-C diagram of V479 Lac reveals a period decrease of d P/dt =-1.69 × 10-6d yr-1. The results of the light curve solutions are:(i) the targets are overcontact binaries with small fill-out factors;(ii) their components are F–K stars, comparable in size, whose temperature differences are below 80 K;(iii) all targets undergo partial eclipses and to limit the possible mass ratios we carried out two-step q-search analysis. The target global parameters(luminosities, radii, masses) were obtained on the basis of their Gaia distances and the results of our light curve solutions. The obtained total mass of V560 Lac turns out to be smaller than the lower mass limit for presently known W UMa binaries of 1.0-1.2 M_⊙, i.e. this target is a peculiar overcontact system.
基金the Human Development Fund for the scholarshipHadhramout University in Yemen for ongoing support
文摘The best physical and geometrical parameters of the main sequence close visual binary system(CVBS), HIP 105947, are presented. These parameters have been constructed conclusively using Al-Wardat’s complex method for analyzing CVBSs, which is a method for constructing a synthetic spectral energy distribution(SED) for the entire binary system using individual SEDs for each component star. The model atmospheres are in its turn built using the Kurucz(ATLAS9) line-blanketed plane-parallel models. At the same time, the orbital parameters for the system are calculated using Tokovinin’s dynamical method for constructing the best orbits of an interferometric binary system. Moreover, the mass-sum of the components, as well as the ?θ and ?ρ residuals for the system, is introduced. The combination of Al-Wardat’s and Tokovinin’s methods yields the best estimations of the physical and geometrical parameters. The positions of the components in the system on the evolutionary tracks and isochrones are plotted and the formation and evolution of the system are discussed.
文摘We present the physical and geometrical parameters of the individual components of the close visual double-lined spectroscopic binary system Gliese 762.1, which were estimated using Al-Wardat＇s complex method for analyzing close visual binary systems. The estimated parameters of the individual components of the system are as follows： radius RA= 0.845±0.09R⊙, RB= 0.795±0.10R⊙, effective temperature Teff^A = 5300luminosity±50 K, Teff^B= 5150 L±50 K, surface gravity log gA= 4.52±0.10, log gB=4.54±0.15 and A= 0.51 with a semi-major axis of 0.0865±0.08L⊙, LB= 0.40±0.07L⊙. New orbital elements are presented±0.010 arcsec using the Hippracos parallax π = 58.96 ndividual masses of the system are determined as M = 1.±0.65 mas, and an accurate total mass and i72±0.60 M⊙,MA= 0.89±0.08 M⊙and MB= 0.83 K0 V and±0.07 M⊙. Finally, the spectral types and luminosity classes of both components are assigned as K1.5V for the primary and secondary components respectively,and their positions on the H-R diagram and evolutionary tracks are given.
文摘多任务学习(Multi-task Learning,MTL)就是把多个问题一起进行分析、计算,以发掘不同问题之间的相关性,提高分析结果的精度,该类方法已被广泛地应用于机器学习、模式识别、计算机视觉等领域.使用多任务学习方案研究了恒星大气物理参数中表面温度(Teff)、表面重力加速度(lg g)、化学丰度([Fe/H])的估计问题.首先使用多任务Sparse Group Lasso算法提取对3个大气物理参数均有预测能力的光谱特征;然后使用支持向量机估计恒星大气物理参数.该方案在Sloan实测恒星光谱和理论光谱上均做了测试.在实测光谱上的平均绝对误差分别为:0.0064(lg(Teff/K)),0.1622(lg(g/(cm·s^(-2)))),0.1221 dex([Fe/H]).在由Kurucz的New Opacity Distribution Function(NEWODF)模型得到的理论光谱上也做了同样的特征提取和恒星大气物理参数估计测试,相应的平均绝对误差分别为:0.0006(lg(Teff/K))),0.0098(lg(g/(cm·s^(-2)))),0.0082 dex([Fe/H]).通过与文献中的同类研究比较表明,多任务Sparse Group Lasso特征提取与支持向量机回归(support vector machine regression,SVR)两者结合的方案有较高的恒星大气物理参量估计精度.
基金supported by the Xinjiang Science Fund for Distinguished Young Scholars under Grant Nos. 2013721014 and 2014721015the National Natural Science Foundation of China under Grant Nos. 11473024, 11363005 and 11163005
文摘Using stellar evolutionary models, we investigate the effects of convective overshooting on naked helium stars. We find that a larger value of overshooting parameter δov results in a larger convective core, which prolongs the lifetimes of naked helium stars on the helium main sequence and leads to higher effective temperatures and luminosities. For naked helium stars with masses lower than about 0.8 Mo, they hardly become giant stars as a result of a weak burning shell. However, naked helium stars with masses between about 0.8 M⊙ and 1.1 M⊙ can evolve into giant branch phases, and finally become carbon oxygen white dwarfs.
