灰色灾变预测模型及其应用

灰色系统是信息不完全系统,不少论文提出了"灰数"、"灰平面"、"灰色统计"等一系列全新概念,为不确定性,信息不完全的数的研究提供了一种新的分析处理方法.灰色方法即是对一些离散的原始数据采用累加生成及其逆进行处理,使之转化为适合用微分方程等方法来建模的有序数列的新方法.本文使用双向差分方法建立了一种灾变预测的灰色预测模型,并对某地区历年来的旱、涝灾数据进行灰色处理,进一步预测到旱、涝灾情将再发生的时间.最后分析了灰色预测模型的拟合精度.     

Influence of the 27-day solar flux variations on the ionosphere parameters measured at Irkutsk in 2003–2005

We present an investigation of the influence of the 27-day solar flux variations, caused by solar rotation, on the ionosphere parameters such as the F2 layer critical frequency (foF2) and the total electron content (TEC). Our observational data were obtained with the Irkutsk Digisonde (DPS-4) located at 52.3 North and 104.3 East during the period from 2003 to 2005. In addition, we use TEC data from the Global Ionosphere Maps (GIM) based on Global Positioning System (GPS) satellites. The solar radiation flux at a wavelength of 10.7 cm (F10.7 index) is used as an index characterizing the solar activity level. A good correlation between observed ionosphere parameters and solar activity variations is found especially in autumn-to-winter season. We estimate the impact of the 27-day solar flux variations on the day-to-day variability and determine the time delay of the ionosphere response.     

基于双层K近邻算法航站楼短时客流量预测

航站楼离港客流量在短时期内呈现准周期性规律变化,易受航班计划、天气等多种因素影响,表现出复杂的非线性特点。为了实现航站楼短时客流量的准确预测,在传统K近邻（KNN）算法基础上增加了航班计划状态模式匹配方法,以航班计划包含的多维属性作为特征选取相似历史运营日作为预测基准向量,建立基于航站楼短时客流量预测的双层K近邻模型。通过实例分析,与ARIMA模型和传统K近邻模型等进行比较,证明双层K近邻模型预测误差更小,精度更高,模型拟合度相对传统K近邻模型提高了8%~10%,为航站楼短时客流量精确预测提供了一种新的解决思路。      

优化的GM(1,1)模型及其适用范围

在已有灰色理论的基础上,利用“最小二乘法”确定GM(1,1)白化权函数的时间响应函数中的常数c,摈弃了传统GM(1,1)把原始序列中x^(0)(1)作为初始条件的欠科学的做法,构建了时间响应函数的优化模型。经大量的数据模拟和预测,发现优化的GM(1,1)模型各项指标均优于传统的GM(1,1)模型,且拓宽了其适用范围。     

地球低纬区边缘大气密度的预测

基于NRLMSISE-00大气模型讨论日地空间环境对地球低纬度地区边缘大气密度的影响,提出预测地球低纬地区边缘大气日平均密度的简化模型法和经验法。简化模型法利用地磁活动和太阳活动的11年准周期特性,通过预测地磁活动和太阳活动的变化规律以预测地球边缘大气密度。经验法则直接利用第23个太阳活动周期的日平均密度变化曲线经过傅里叶变换处理得到日平均密度变化规律曲线,然后将曲线拟合得到不同高度下的密度昼夜波动规律,再利用预测得到的日平均密度即可计算出具体当地时间对应的密度情况。误差分析说明经验预测法比简化模型法精度更高。两种方法均具有较高的精度并且使用方便,可用于地球边缘大气密度的工程化预测计算。     

基于ARIMA模型的拉萨降水量时间序列分析与预测

应用差分自回归滑动平均模型（ARIMA）模型对西藏拉萨地区1969年至2009年的年降水量资料进行建模分析。建立的模型通过了参数的显著性检验和模型的显著性检验。模型的实证分析表明在短期内模型具有较高的预测精度,ARIMA模型可以较好的应用于降水量的预测,对相关部门采取措施应对自然灾害提供了理论支持。     

基于GM(1,1)分段组合的宽体货机预测模型研究

航空货运市场是全球航空运输市场的重要组成部分,与全球经济、贸易的发展息息相关。航空货运市场的发展趋势及需求直接影响到货机市场的发展趋势及货机行业的发展规划。而作为全球航空货运市场的运输载体,货机尤其是远程宽体货机对货物的国际间运输和交流起到了至关重要的作用。当前主流的预测方法为根据需求与供应的对应关系进行预测,预测过程相对复杂,预测的精度相对较高,不足之处在于预测工作周期较长,难以快速获得预测的结果。根据当前灰色系统理论,建立了宽体货机机队预测的灰色预测模型GM(1,1),并用分段组合的方式,较为精确地预测宽体货机未来机队的发展情况,预测较为快速,效率较高,能满足一般精度的预测需求。     

未来太空气象预报系统一瞥

文章从几个方面对太空气象预报系统进行了简单的介绍,包括:当前太空气象预报系统的水平;模板预报方案实施细则;实施模板方案的难度与可行性。作者关心我国对太空气象预报系统的研究,为进一步降低航天器发射风险、提高航天器运行可靠性而献计献策。     

Space weather center in Japan

Progress in information technology has enabled to collecting data in near real-time. This significantly improves our ability to monitor space weather conditions. We deliver information on near real-time space weather conditions via the internet. We have started two collaborations with space weather users. One is a measurement of geomagnetically induced current (GIC) of power grids in collaboration with a Japanese power company. The other concerns radiation hazards for aircrews. The radiation exposure level for aircrews was been determined by the Japanese government by the end of 2005. The proposed upper limit is 5 mSV a year. We are actively seeking ways to contribute to this subject. Our activities at the Japanese space weather center are reported in this paper.     

Rigidity spectrum of the 27-day variation of the galactic cosmic ray intensity in different epochs of solar activity

We study the temporal evolution of the power rigidity spectrum of the first (27 days) and the second (14 days) harmonics of the 27-day variation of the galactic cosmic ray intensity measured by neutron monitors in the period of 1965–2002. The rigidity spectrum of these variations can be approximated by a power law. We show the rigidity spectra of the first and the second harmonics of the 27-day variation of the galactic cosmic ray intensity have similar time profiles. These spectra are hard (γ ≈ 0.5 ± 0.1) and soft (γ ≈ 1.1 ± 0.2) during solar maximum and minimum activity, respectively. We ascribe this to the alternation of the sizes of the modulation regions responsible for the 27-day variation of the galactic cosmic ray intensity in different epochs of solar activity. Especially, the average radial sizes of the modulation regions of the 27-day variation and the heliolatitudinal extension of the heliolongitudinal asymmetry are smaller during solar minimum than during solar maximum. We show also, that the temporal changes of the power rigidity spectra of the first and the second harmonics of the 27-day variation of the galactic cosmic ray intensity are in a negative correlation with the changes of the rigidity spectrum of the corresponding 11-year variation.