随着卫星技术的飞速发展,保障空间技术系统的运行安全对空间环境预报业务提出了更高要求.本文根据空间环境信息服务中数据及模型资源分布式集成与共享的现状,针对空间环境预报数据智能化、自动化生成的需求,将Web服务技术及语义服务组合技术应用到空间环境信息服务中.根据空间环境领域特点,提出了空间环境领域的语义相似度计算方法,并采用基于语义相似度的方法实现空间环境信息服务的自动组合.

日冕物质抛射(Coronal Mass Ejection,CME)参数识别模型是太阳风预报过程的重要组成部分.在空间环境预报业务中,为提高太阳风预报的准确率,需要提高CME参数识别的精度.模型以计算任务串行的方式运行,运算效率低导致模型运算时间长,不能满足这种需求.CME参数识别模型的物理运算过程相互不独立,其在单节点上的运行方式不能满足并行化要求.基于MapReduce的并行计算框架,改进了CME参数识别模型的计算流程,提出CDMR (CME detection under MapReduce)方法,实现了CME参数识别模型的并行计算,并对比分析CME参数识别模型在串行计算和MapReduce并行计算下的运行时间,提高了模型的识别精度和计算效率.

空间环境监测网需要通过广域网将海量空间环境数据从地面监测台站传输到数据中心.本文研究了TCP拥塞控制机制,采用BBR拥塞控制算法解决数据传输存在的问题.针对空间环境数据传输需求,设计了空间环境数据传输的工作流模型,建立了空间环境数据传输系统.实验证明该系统适于进行广域网链路上的空间环境数据传输.

Ionospheric data assimilation is a now-casting technique to incorporate irregular ionospheric measurements into certain background model, which is an effective and efficient way to overcome the limitation of the unbalanced data distribution and to improve the accuracy of the model, so that the model and the data can be optimally combined with each other to produce a more reliable and reasonable system specification. In this study, a regional total electron content (TEC) now-casting system over China and adjacent areas (70 degrees E 140 degrees E and 15 degrees N -55 degrees N) is developed on the basis of data assimilation technique. The International Reference Ionosphere (IRI) is used here as background model, and the GNSS data are derived from both the Space Environment Monitoring Network of Chinese Academy of Sciences (SEMnet) and International GNSS Service (IGS) data. A Three-dimensional variation algorithm (3DVAR) combined with Gauss-Markov Kalman filter technique is used to implement the data assimilation. The regional gridded TEC maps and the position errors of single-frequency GPS receivers can be generated and publicized online (http://sepc.ac.cn/TEC_chn.php) in quasi-real time, which is updated for every 15 min. It is one of the ionospheric now casting systems in China based on data assimilation algorithm, which can be used not only for real-time monitoring of ionosphere environment over China and adjacent areas, but also in providing accurate and effective specification of regional ionospheric TEC and error correction for satellite navigation, radar imaging, shortwave communication, and other relevant applications.

With the development of the aerospace industry,the space environment may cause more and more serious harm to the space activities. In order to protect the normal work and extend the service life of spacecraft,the space environment research needs to provide fast,comprehensive and applied space environment information. The existing space environment research mode is complicated in operation process, low in operation efficiency, scattered in quantity and independent in operation and disadvantageous in management. In this paper,an integrated platform for business-oriented spatial environment mode based on Petri nets is proposed,which can be easily integrated and united to meet the needs of business operation through code-free integration and task scheduling. The platform realizes the friendly interface integrated configuration and 7 * 24 operation,convenient and unified management of multi-mode,enhance the spatial environment mode of business level.

With the increasing of users and the demands which are transforming from the monotonous traditional service to network service,Multiple space environment web applications including browser-client,rich-client and mobile-client applications have been developed by SEPC（Space Environment Prediction Center,NSSC,CAS）during the past few years.The architecture of Operational Space Environment Technology System（OSETS） that these applications rely on is described and the description of structural optimization of the architecture is provided.To demonstrate the evolution of the OSETS,three web application examples for e SpaceWx,Space Weather Situation Awareness Picture（SWSAP）,Plug-and-Play SWx Analysis and Plotting Program（PPSWAP） are presented.

针对数据源复杂、实时性强、准确性高和数据类型多样的Web空间环境数据采集任务,提出了一个基于Petri网的信牌驱动式Web数据采集模型.首先,通过引入Petri网的基本要素作为模型的理论基础,研究适合于Web数据采集的建模方法;在此基础上,针对模型的具体应用验证,研究了空间环境数据采集任务服务系统(SEDGSS)的架构设计,对数据源配置子系统、任务控制子系统和任务处理子系统进行具体的实现.实验结果表明,该模型实现了自动化机制和回溯校验机制,并具有良好的易配置性、可重用性和扩展灵活性;该系统7×24小时实时抓取254个复杂的数据源任务,目前正承担着自动化、业务化的空间环境数据采集任务以服务于我国空间环境预报.

剧情是太阳风暴应急演练的驱动,控制演练的过程和发展。针对传统演练中手动设置剧情、演练剧情已知,且不能反映太阳风暴动态演化过程等问题,利用二叉树重构法建立太阳风暴剧情推理引擎,根据不同演练需求智能化地生成太阳风暴剧情方案;基于历史事件库,利用分段线性插值等方法,真实表现了难以用物理模型描述的太阳风暴动态演化过程;基于以上两项关键技术,实现了太阳风暴剧情生成系统。实验结果表明,系统生成的太阳风暴剧情,贴近实战,针对性强,可有效提升演练中各部门对突发事件的快速反应和应急处置能力。