CMOS图像传感器在太阳磁场观测中的应用研究

磁场是太阳物理的第1观测量,当前太阳磁场观测研究正迈向大视场、高时空分辨率、高偏振测量精度以及空间观测的时代.中国首颗太阳观测卫星—先进天基太阳天文台(ASO-S)也配置了具有高时空分辨率、高磁场灵敏度的全日面矢量磁像仪(FMG)载荷,针对FMG载荷的需求,讨论了大面阵、高帧频互补金属氧化物半导体(Complementary Metal Oxide Semiconductor, CMOS)图像传感器应用于太阳磁场观测的可行性.首先,基于滤光器型太阳磁像仪观测的原理,比较分析了目前CMOS图像传感器(可用的或是可选的两种快门模式)的特点,指出全局快门类型更适合FMG;其次搭建了CMOS传感器实验室测试系统,测量了CMOS图像传感器的像素增益及其分布规律;最后在怀柔太阳观测基地的全日面太阳望远镜上开展了实测验证,获得预期成果.在这些研究基础上,形成了FMG载荷探测器选型方向.     

基于边缘探测器的FMG稳像探测研究

全日面矢量磁像仪(Full-disk vector Magnetograph, FMG)是先进天基太阳天文台(Advanced Space-based Solar Observatory, ASO-S)卫星的主要载荷之一,为确保其可获得高空间分辨率的全日面太阳矢量磁场图, FMG配备了一套高精度稳像控制系统.FMG稳像系统采用边缘探测器对太阳入射倾斜偏差进行测量.基于FMG望远镜的太阳像仿真数据分析边缘探测器在太阳像能量分布下的输出特性以及太阳像尺寸与边缘探测器量程和灵敏度的关系.最后,在实验室搭建了一套验证系统对FMG稳像系统边缘探测器进行了相关性能验证实验.实验结果与理论分析一致,太阳像尺寸越大,边缘探测器量程越大,边缘探测器灵敏度越低.     

基于SIFT特征检测和密度峰值聚类的太阳活动区自动检测算法研究

太阳活动区是太阳大气中产生各种活动现象的区域,精确地检测和识别太阳活动区对理解太阳磁场的形成机制具有极为重要的科学意义.根据太阳活动区结构较为复杂的特点,基于尺度不变特征变换(ScaleInvariant Feature Transform, SIFT)和密度峰值聚类(Clustering by Fast Search and Find of Density Peaks,DPC)算法的优越性,提出了一种太阳活动区的自动检测和识别方法.首先,对太阳动力学天文台(Solar Dynamics Observatory, SDO)日震和磁场成像仪(Helioseismic and Magnetic Imager, HMI)的纵向磁图进行对比度增强;然后采用SIFT方法提取出全日面磁图中的特征点;最后利用DPC算法将特征点进行聚类,从而自动检测和识别出太阳活动区.研究结果表明, SIFT和DPC算法相结合的方法可以在不需要人工交互的情况下准确地自动检测出太阳活动区.     

FMG载荷地面试观测导行跟踪系统的设计与实现

全日面矢量磁像仪(Full-disk vector MagnetoGraph, FMG)是先进天基太阳天文台(Advanced Space-based Solar Observatory, ASO-S)卫星的3台主载荷之一,为开展FMG全系统性能测试和定标试验,已搭建用于FMG外场测试的地面试观测平台.利用该平台模拟FMG在轨跟踪状态,研制了基于全日面太阳图像的望远镜导行系统.该系统通过大面阵CCD (Charge Coupled Device)采集太阳像、多重逻辑条件判定、微调恒动跟踪速度校正偏移等策略,实现了RMS (Root Mean Square)优于1′′/30 min的跟踪精度.通过分析FMG方案阶段试观测的太阳纵向磁图,开启导行30 min后磁图特征点在赤经方向的偏移比恒动条件下减少17.5′′,提升了磁图空间分辨率.测试过程中该系统达到设计指标且工作稳定,为FMG地面试观测提供了良好的技术支撑.     

太阳表面磁场的分布与演化研究进展

太阳磁场、较差自转和内部对流使得日面磁场与磁活动在很大的时间尺度和空间尺度范围均表现得相当复杂.其中最有名的是太阳活动的11年周期,或22年磁周期.在较小时间尺度上,从几秒到几小时,有时太阳大气中会发生一些壮观的爆发事件,如耀斑、日珥爆发、日冕物质抛射等.所有这些形式的事件都与太阳磁场紧密关联.简单评述了太阳磁场起源与观测方法,重点论述了不同尺度太阳磁场的空间分布与演化,介绍了从太阳磁活动现象统计得到的有关太阳磁场的几个典型特征,同时讨论了进一步研究的方向.     

太阳Stokes光谱望远镜的结构及其数据处理方法

云南天文台的太阳Stokes光谱望远镜是一台通过测量磁敏谱线的Stokes参数I,Q,U和V的轮廓来研究太阳磁场精细结构的光谱型矢量磁场测量仪。它利用4个完整的Stokes轮廓所蕴含的丰富信息,完全确定辐射的偏振状态,从而精确地测定太阳黑子区的矢量磁场。文章首先介绍了该望远镜的结构,进而详细地介绍了该望远镜所测量的偏振光谱资料的处理方法。     

太阳活动极大和极小期太阳磁场周期变化研究

利用Wilcox天文台1975年到2010年间的太阳磁场数据,分析了太阳平均磁场在太阳活动极大和极小时期的短时周期性.结果显示太阳磁场主要具有9 d、13.5 d、27 d左右的周期.在太阳活动极大时期,27 d左右周期最为显著,而在太阳活动极小时期最显著的周期为13.5 d左右(1984～1986年间的太阳活动极小时期除外).这些结果说明太阳的活动区域在活动极大和极小时期具有明显不同的分布.     

新浮磁流与增亮事件

本文利用阳光卫星的软Ｘ 射线资料,日本国立天文台的Ｈα单色像和相对应的北京天文台怀柔站的活动区矢量磁场资料及美国基特峰天文台的全日面磁图,对发生在１９９４ 年５ 月１８ 日的软Ｘ 射线和Ｈα增亮事件进行了仔细的研究。研究表明增亮事件明显与新浮剪切磁流出现相关,Ｈα和软Ｘ 射线亮点的投影位置正好位于新浮小磁流环的三个足点之间,同时亮点也位于分界面与光球面的交线附近,故该增亮事件也许是太阳低层大气磁重联的观测证据     

基于改进VNet太阳暗条检测方法

太阳暗条作为太阳大气磁场的示踪,对研究太阳磁场有极其重要的意义。针对现有的暗条检测方法存在检测精度不高,弱小暗条错检、漏检等问题,提出一种基于改进VNet网络的太阳暗条检测方法。首先,使用大熊湖天文台Hα全日面图像并结合磁图制作了太阳暗条数据集;其次,在VNet网络下采样部分采用Inception模块融合不同尺度特征图的特征,同时加入注意力机制增强特征图中暗条部分的语义信息;最后在上采样部分引入深度监督模块,更多地保留太阳暗条的细节特征。为验证算法性能,采用191幅Hα全日面图像数据集,其中包含暗条共3372条。算法在测试数据集上平均准确率达到0.9883,F1值达到0.8385。实验结果证明,该方法可以有效识别Hα全日面图中的暗条。     

1981年4月1日日面4N耀斑环系的光学前兆与活动特征

根据云南天文台太阳色球H_α和相应的光球黑子观测以及磁场测量,并结合有关X—ray资料等,对1981年4月1日日面4N大耀斑进行了部份测量和分析。结果表明,该耀斑为环系;光球浮现磁流和黑子扭曲、挤压和剪切运动是触发该耀斑的直接原因。而活动的黑子光桥又是浮现磁流的一种重要标志;耀斑环或带与磁场位形密切相关;耀斑后在运动空间原位置处光球又浮现出部份磁流;卵形暗条内预示能爆发大耀斑。     

Irreversible rapid changes of magnetic field associated with the 2012 October 23 circular near-limb X1.8 Flare

It has been found that photospheric magnetic fields can change in accordance with restructuring of the three-dimensional magnetic field following solar eruptions.Previous studies mainly use vector magnetic field data taken for events near the disk center.In this paper,we analyze the magnetic field evolution associated with the 2012 October 23 X1.8 flare in NOAA AR 11598 that is close to the solar limb,using both the 45 s cadence line-of-sight and 12 min cadence vector magnetograms from the Helioseismic and Magnetic Imager on board Solar Dynamics Observatory.This flare is classified as a circular-ribbon flare with spine-fan type magnetic topology containing a null point.In the line-of-sight magnetograms,there are two apparent polarity inversion lines(PILs).The PIL closer to the limb is affected more by the projection effect.Between these two PILs there lie positive polarity magnetic fields,which are surrounded by negative polarity fields outside the PILs.We find that after the flare,both the apparent limb-ward and disk-ward negative fluxes decrease,while the positive flux in-between increases.We also find that the horizontal magnetic fields have a significant increase along the disk-ward PIL,but in the surrounding area,they decrease.Synthesizing the observed field changes,we conclude that the magnetic fields collapse toward the surface above the disk-ward PIL as depicted in the coronal implosion scenario,while the peripheral field turns to a more vertical configuration after the flare.We also suggest that this event is an asymmetric circular-ribbon flare:a flux rope is likely present above the disk-ward PIL.Its eruption causes instability of the entire fan-spine structure and the implosion near that PIL.     

ASO-S卫星HXI量能器探测单元的标定

先进天基太阳天文台卫星(Advanced Space-based Solar Observatory, ASO-S)是中国科学院第2批空间科学先导专项之一,其主要目标是同时观测太阳磁场、耀斑和日冕物质抛射,并对3者之间的相互关系和内在联系进行研究.硬X射线成像仪(HXI)是ASOS卫星的3大载荷之一,它通过对太阳活动发射的硬X射线进行傅里叶调制成像,实现高空间分辨率和高时间分辨率的太阳能谱成像观测.量能器单机是HXI的关键单机之一,其主要任务是精准测量通过每对光栅后太阳硬X射线的能量和通量.主要介绍了量能器单机的工作原理及其关键指标要求、标定设备及标定方案,最后给出了标定结果,从而验证了量能器单机方案设计的合理性.     

空间太阳硬X射线成像仪量能器读出电子学设计

先进天基太阳天文台(ASO-S)是中国科学院空间科学先导专项2期规划的太阳观测卫星,其针对第25个太阳活动峰年,同时观测太阳磁场、日冕物质抛射和太阳耀斑爆发.硬X射线成像仪(HXI)作为该卫星3个科学载荷之一,实现了高时间分辨率和空间分辨率的太阳硬X射线成像观测,其量能器由99套溴化镧闪烁晶体-光电倍增管探测单元和读出电子学板构成,实现了30–200 keV的硬X射线光子能谱测量.针对HXI量能器的观测需求,设计了一套空间高事例率读出电子学系统,并通过实验室测试,证明了该系统单事例读出死时间小于2μs,同时验证了该系统电子学噪声小于120 fC,积分非线性小于2%,满足HXI仪器要求.     

Magnetic Field Vector Retrieval With the Helioseismic and Magnetic Imager

We investigate the accuracy to which we can retrieve the solar photospheric magnetic field vector using the Helioseismic and Magnetic Imager (HMI) that will fly onboard of the Solar Dynamics Observatory by inverting simulated HMI profiles. The simulated profiles realistically take into account the effects of the photon noise, limited spectral resolution, instrumental polarization modulation, solar p modes, and temporal averaging. The accuracy of the determination of the magnetic field vector is studied by considering the different operational modes of the instrument.     

RECONSTRUCTION OF THE THREE-DIMENSIONAL CORONAL MAGNETIC FIELD

Studies of solar flares indicate that the mechanism of flares is magnetic in character and that the coronal magnetic field is a key to understanding solar high-energy phenomena. In our ongoing research we are conducting a systematic study of a large database of observations which includes both coronal structure (from the Soft X-ray Telescope on the Yohkoh spacecraft) and photospheric vector magnetic fields (from the Haleakala Stokes Polarimeter at Mees Solar Observatory). We compare the three-dimensional nonlinear force-free coronal magnetic field, computed from photospheric boundary data, to images of coronal structure. In this paper we outline our techniques and present results for active region AR 7220/7222. We show that the computed force-free coronal magnetic field agrees well with Yohkoh X-ray coronal loops, and we discuss the properties of the coronal magnetic field and the soft X-ray loops.     

DeepSun:machine-learning-as-a-service for solar flare prediction

Solar flare prediction plays an important role in understanding and forecasting space weather.The main goal of the Helioseismic and Magnetic Imager(HMI), one of the instruments on NASA's Solar Dynamics Observatory, is to study the origin of solar variability and characterize the Sun's magnetic activity.HMI provides continuous full-disk observations of the solar vector magnetic field with high cadence data that lead to reliable predictive capability; yet, solar flare prediction effort utilizing these data is still limited. In this paper, we present a machine-learning-as-a-service(MLaa S) framework, called Deep Sun,for predicting solar flares on the web based on HMI's data products. Specifically, we construct training data by utilizing the physical parameters provided by the Space-weather HMI Active Region Patch(SHARP)and categorize solar flares into four classes, namely B, C, M and X, according to the X-ray flare catalogs available at the National Centers for Environmental Information(NCEI). Thus, the solar flare prediction problem at hand is essentially a multi-class(i.e., four-class) classification problem. The Deep Sun system employs several machine learning algorithms to tackle this multi-class prediction problem and provides an application programming interface(API) for remote programming users. To our knowledge, Deep Sun is the first MLaa S tool capable of predicting solar flares through the internet.     

Vertical current and a 3B/X12 flare in a highly sheared active region (NOAA 6659) on June 9, 1991

NOAA active region 6659, during its June 1991 transit across the solar disk, showed highly sheared vector magnetic field structures and produced numerous powerful flares, including five white-light flares. Photospheric vector magnetograms of this active region were obtained at the Huairou Solar Observing Station of the Beijing Astronomical Observatory. After the resolution of the 180° ambiguity of the transverse magnetic field and transformation of off-center vector magnetograms to the heliographic plane, we have determined the photospheric vertical current density and discussed the relationship with powerful flares. The following results were obtained: (a) The powerful 3B/X12 flare on June 9, 1991 was triggered by the interaction between the large-scale electric current system and magnetic flux of opposite polarity. (b) The kernels of the powerful Hβ flare (sites of the white-light flare) were close to the peaks of the vertical electric current density. (c) Some small-scale structures of the vertical current relative to the magnetic islands of opposite polarity have not been found. This probably implies that the electric current is not always parallel to the magnetic field in solar active regions.     

ASO-S/HXI太阳指向镜算法研究

硬X射线成像是研究太阳耀斑等爆发现象的重要手段.由于采用调制成像而非直接成像的原因, X射线图像在日面上的位置需要借助太阳指向镜提供的仪器指向的日面坐标来确定.因此,指向信息对于耀斑定位实现多波段研究,理解太阳耀斑的物理过程具有重要的科学意义.在此对两种太阳指向镜指向信息的获取算法进行了测试.结合太阳指向镜的设计方案,首先利用SDO (Solar Dynamics Observatory)/AIA (Atmospheric Imaging Assembly) 4500?的数据产生测试图像,其次对其进行二值化处理,分别提取日面轮廓和4个边角指定区域面积;最后分别利用最小二乘法和四象限法对太阳中心坐标进行反演.初步结果显示最小二乘法受随机噪声影响小,定位精度相对稳定约为0.25′′,并可提供四象限法解算的初值;后者的精度可以优于0.14′′,但受随机噪声影响较大.两种算法的精度都显著优于硬X射线成像仪(Hard X-ray Imager, HXI)太阳指向镜的设计要求,可为指向数据在将来科学分析中的实际应用提供参考.