基于混沌时间序列的Volterra自适应滤波极移预报方法

针对极移复杂的时变特性, 根据混沌相空间坐标延迟重构理论, 提出一种基于Volterra自适应滤波的极移预报方法. 首先, 利用最小二乘拟合算法分离极移序列中的线性趋势项、钱德勒项和周年项, 获得线性极移、钱德勒极移和周年极移的外推值; 其次, 通过C-C关联积分法对最小二乘拟合残差序列进行相空间重构, 并利用小数据量法计算残差序列的最大Lyapunov指数验证其混沌特性, 在此基础上, 构建Volterra自适应滤波器对残差序列进行预测; 最后, 将线性极移、钱德勒极移和周年极移的外推值以及最小二乘拟合残差的预测值相加获得极移最终预报值. 利用国际地球自转服务局(International Earth Rotation and Reference Systems Service, IERS)提供的极移数据进行1--60d跨度预报, 并将预报结果分别与国际地球定向参数预报比较竞赛(Earth Orientation Parameters Prediction Comparison Campaign, EOP PCC)结果和IERS A公报发布的极移预报产品进行对比, 结果表明: 对于1--30d的短期预报, 该方法的预报精度与EOP PCC最优预报方法相当, 当预报跨度超过30d时, 该方法的预报精度低于EOP PCC最优预报方法, 优于参与EOP PCC的其他方法; 与IERS A公报相比, 该方法的短期预报效果较好, 当预报跨度增加时预报精度低于IERS A公报. 预报结果表明该方法更适合于极移短期预报.     

基于有效角动量函数的多参数极移预报

极移的变化与多种激发息息相关,这些激发包括大气表面压力和大气风、海底压力和洋流、陆地水分布以及气候变暖导致的海平面变化,并且可以通过有效角动量函数来估计.在极移预报中,通过刘维尔方程融合有效角动量函数,并利用最小二乘与自回归组合的方法进行拟合及外推,同时,对自回归模型的可调节参数设置更多的选择,在不同的极移预报阶段,对于不同分量的预报匹配更优的参数,有效地提高了极移的预报精度.在441次1–90 d的极移预报实验中,短中期的预报改善更为明显,在1–6 d和7–30 d的极移X预报结果中,分别有56.9%和53.5%优于国际地球自转服务(International Earth Rotation Service, IERS)的预报;在1–6 d和7–30 d的极移Y预报结果中,分别有66.5%和59.7%优于IERS的预报.整体上,极移Y的预报精度比极移X的预报精度有更多的提升,以IERS的地球定向参数(Earth Orientation Parameters, EOP)产品EOP 14C04 (IAU2000A)为参考,极移X预报在第1 d、第5 d的MAE (Mean Absolute...     

考虑LS+AR模型基础数据量的极移预报研究

针对目前极移最小二乘(Least Square, LS)+自回归(AutoRegressive, AR)预报模型的单一数据选取方案, 提出分别考虑LS模型数据量和AR残差数据量的组合数据模式, 并对极移预报时单一数据和组合数据预报结果精度进行分析, 探讨模型输入数据量对极移预报精度的影响. 结果表明, 模型输入数据量的变化对极移预报结果影响较大. 采用组合数据预报的方式相比较于单一数据量预报方式精度更高, 特别是针对30--360 d跨度内的中长期预报, 组合数据量的极移预报精度可比单一数据量预报精度有较大改善. 结论证明组合数据在极移预报时具有一定的优势, 可为以后极移预报数据量选取提供一定的借鉴参考意义.     

A New Prediction Method for the Arrival Time of Interplanetary Shocks

Solar transient activities such as solar flares, disappearing filaments, and coronal mass ejections (CMEs) are solar manifestations of interplanetary (IP) disturbances. Forecasting the arrival time at the near Earth space of the associated interplanetary shocks following these solar disturbances is an important aspect in space weather forecasting because the shock arrival usually marks the geomagnetic storm sudden commencement (SSC) when the IMF Bz component is appropriately southward and/or the solar wind dynamic pressure behind the shock is sufficiently large. Combining the analytical study for the propagation of the blast wave from a point source in a moving, steady-state, medium with variable density (wei, 1982; wei and dryer 1991) with the energy estimation method in the ISPM model (smith and dryer 1990, 1995), we present a new shock propagation model (called SPM below) for predicting the arrival time of interplanetary shocks at Earth. The duration of the X-ray flare, the initial shock speed and the total energy of the transient event are used for predicting the arrival of the associated shocks in our model. Especially, the background speed, i.e., the convection effect of the solar wind is considered in this model. Applying this model to 165 solar events during the periods of January 1979 to October 1989 and February 1997 to August 2002, we found that our model could be practically equivalent to the prevalent models of STOA, ISPM and HAFv.2 in forecasting the shock arrival time. The absolute error in the transit time in our model is not larger than those of the other three models for the same sample events. Also, the prediction test shows that the relative error of our model is ≤10% for 27.88% of all events, ≤30% for 71.52%, and ≤50% for 85.46%, which is comparable to the relative errors of the other models. These results might demonstrate a potential capability of our model in terms of real-time forecasting.     

A New Method for Polar Field Interpolation

The photospheric magnetic field in the Sun’s polar region is not well observed compared to the low-latitude regions. Data are periodically missing due to the Sun’s tilt angle, and the noise level is high due to the projection effect on the line-of-sight (LOS) measurement. However, the large-scale characteristics of the polar magnetic field data are known to be important for global modeling. This report describes a new method for interpolating the photospheric field in polar regions that has been tested on MDI synoptic maps (1996 – 2009). This technique, based on a two-dimensional spatial/temporal interpolation and a simple version of the flux transport model, uses a multi-year series of well-observed, smoothed north (south) pole observations from each September (March) to interpolate for missing pixels at any time of interest. It is refined by using a spatial smoothing scheme to seamlessly incorporate this filled-in data into the original observation starting from lower latitudes. For recent observations, an extrapolated polar field correction is required. Scaling the average flux density from the prior observations of slightly lower latitudes is found to be a good proxy of the future polar field. This new method has several advantages over some existing methods. It is demonstrated to improve the results of global models such as the Wang–Sheeley–Arge (WSA) model and MHD simulation, especially during the sunspot minimum phase.     

基于自适应滤波的干扰消除方法研究

射电天文信号非常微弱,电磁环境对射电望远镜观测至关重要,通常可以利用地形、建立无线电宁静区、进行电磁屏蔽与防护等手段来减小电磁干扰.然而,仍有一些干扰难以屏蔽.故提出了一种基于自适应滤波的干扰消除方法,可用于复杂噪声环境中天文信号的提取.该方法借助自适应横向滤波器,采用最小均方(Least Mean Square, LMS)误差算法,以系统误差和收敛性为评判标准,通过改变步长与阶数对滤波效果进行优化,仿真结果显示该滤波器能在保证算法收敛的前提下有效提取信号.为了检验该算法的有效性,选取了新疆天文台南山26 m射电望远镜和Parkes 64 m射电望远镜记录的观测数据,采用设计的滤波器分别对不同的实测数据进行测试,验证了该滤波器的有效性.理论分析与实验结果一致表明该方法能有效消除天文观测中的干扰信号,具有一定的实用性.     

Real-time Solution of Atomic Clock Operating Parameters Based on Adaptive Kalman Filtering

The real-tine and accurate knowledge of the operation of an atomic clock is of great significance to satellite navigation and time keeping. In the light of the problems which exist in the real-time solution of the atomic clock operating parameters by means of the standard Kalman filtering, the method for applying the adaptive Kalman filtering model based on the prediction residual to the real-time solution of the clock parameters is proposed. An adaptive factor is constructed by the model based on the standard Kalman filtering model so that the accurate and reliable estimates of the operating parameters may be given rather fast after the frequency modulation or phase modulation of the atomic clock. Finally, in combination with the 3 computational examples of the work on the navigation and time keeping, the effectiveness of the method is verified.     

Application of Vondrak Filtering Method to the Two-Way Satellite Time and Frequency Transfer Based on Software Defined Receiver

The SDR-TWSTFT (Two-Way Satellite Time and Frequency Transfer based on Software Defined Receiver) link collects measurement data per second and then fits the raw data into 300 seconds by mathematical model. The time transfer results of the link are therefore affected by short-term measurement noise and non-model errors, and exhibit certain characteristics of random noise. A frequency domain amplitude analysis method is proposed to determine the filtering factor and construct a low-pass Vondrak filter to meet the requirements. By analyzing the measured data of the SDR-TWSTFT link between the NTSC (National Time Service Center) of Chinese Academy of Sciences and the PTB (Physikalisch-Technische Bundesanstalt) of Germany, it is found that the method is effective in filtering the high frequency noise in the average time of the link, can improve the confidence of the link time transfer results, and also the short-term frequency and time stability of the filtered link is significantly improved.     

A New Method for Prediction of Peak Sunspot Number and Ascent Time of the Solar Cycle

The purpose of the present study is to develop an empirical model based on precursors in the preceding solar cycle that can be used to forecast the peak sunspot number and ascent time of the next solar cycle. Statistical parameters are derived for each solar cycle using “Monthly” and “Monthly smoothed” (SSN) data of international sunspot number (R _i). Primarily the variability in monthly sunspot number during different phases of the solar cycle is considered along with other statistical parameters that are computed using solar cycle characteristics, like ascent time, peak sunspot number and the length of the solar cycle. Using these statistical parameters, two mathematical formulae are developed to compute the quantities [Q _C] _n and [L] _n for each nth solar cycle. It is found that the peak sunspot number and ascent time of the n+1th solar cycle correlates well with the parameters [Q _C] _n and [L] _n /[S _Max] _n+1 and gives a correlation coefficient of 0.97 and 0.92, respectively. Empirical relations are obtained using least square fitting, which relates [S _Max] _n+1 with [Q _C] _n and [T _a] _n+1 with [L] _n /[S _Max] _n+1. These relations predict a peak of 74±10 in monthly smoothed sunspot number and an ascent time of 4.9±0.4 years for Solar Cycle 24, when November 2008 is considered as the start time for this cycle. Three different methods, which are commonly used to define solar cycle characteristics are used and mathematical relations developed for forecasting peak sunspot number and ascent time of the upcoming solar cycle, are examined separately.     

A Research on the Pulsar Timing Based on Kalman Filtering

The pulsar is a high-speed rotationing neutron star with a stable rotational period, being not disturbed and destroyed artificially, and can be taken as the reference quantity of the absolute time. In this article a kind of pulsar time service method based on the Kalman filtering algorithm is proposed, and the simulation analysis of the clock error control based on the Kalman filtering and of the effect of the pulsar catalogue error and the measuring accuracy of the pulsar time of arrival (TOA) on the accuracy of time service is made by taking a certain solar synchronous orbit as an example. The result shows that by utilizing this method the clock error of the satellite-borne clock can be effectively eliminated and its time-dependent increase is restrained, thereby solving the problem that the accuracy of the spacecraft-borne low cost clock can not meet the needs.     

The Construction of the Theory of Motion for Solar-System Bodies Based on a Universal Method for the Perturbative Function Calculation

A new universal method for series expansion of the perturbative function is developed. A unified theory of motion for Solar-System bodies is constructed on its basis. In the first part of the work, the right-hand sides of equations are limited by the first-order terms.     

双通道终端进行卫星双向法时间比对的归算方法

用双通道终端进行卫星双向法时间比对,可同时实现多台站时间同步。但是,目前双通道终端卫星双向法时间比对仍采用经典二台站归算方法,同时性的优点并没有得到充分的利用,其结果不闭合,比对得到的结果不统一。为了避免了上述问题,给出了新的归算方法,在同一系统下给出时间比对结果。新的归算方法不但提高了时间比对精度,同时还给出卫星到地面站之间的精确伪距,籍此精确地测定卫星的实时位置。     

一种新的守时型原子钟综合时间尺度方法研究

在这篇文章中, 提出了一种基于改进的指数平滑和Vondrak_Cepek联合平滑的氢铯综合时间尺度产生方法. 以最小误差方法为理论基础, 动态估计氢原子钟频率漂移参数, 提升氢原子钟钟差预测准确度; 基于改进的二次指数平滑产生氢原子钟组时间尺度、加权平均方法产生铯原子钟时间尺度, 同时设计Vondrak_Cepek滤波器以结合两类时间尺度长短期稳定度优势, 提升综合时间尺度性能. 实验结果表明, 所提方法产生的氢铯综合时间尺度时稳可达1.60x10^-15,天稳可达3x10^-15,优于ALGOS、AT1和Kalman滤波3种经典方法产生的时间 尺度性能.     

A Method for the Construction of a Restricted Motion in an Arbitrary Central Field

The well-known problem of motion in a central field integrable in quadratures is considered. The force function of the problem depends only on the particle distance to the chosen coordinate origin. In the general case of an arbitrary central force, a rigorous analytical solution of the problem cannot be obtained due to the complexity of the integrals. In this paper we propose a semi-analytical method of constructing an approximate solution for the case where the distance varies in a limited range that allows the time dependences of the polar coordinates to be obtained using elliptic functions and integrals. As an example, we consider the model problems of the perturbed motion of hypothetical Jovian and lunar equatorial satellites as well as the problem of the motion of a single star in the principal plane of a galaxy. The methodical accuracy has been estimated by a comparison with the numerical solution.     

A New Method for Processing the Correlation of Light Variation Time Delays

The method of time delay correlation function (TDCF) is a new method for calculating the sequential time delays. This method is utilized to calculate the time delays in 3 radio wave bands (4.8 GHz, 8 GHz and 14.5 GHz) of the blazar 0316 + 413 (NGC 1275) and to make the multi-band correlation analysis for other 7 blazar sources. The calculated result of the blazar 0316 + 413 shows that its light curve at 4.8 GHz lags by 410 d behind that at 8 GHz, or τ_4.8−8 = 410 d; the light curve at 4.8 GHz lags by 440 d behind that at 14.5 GHz, or τ_4.8−14.5 = 440 d; and the light curve at 8 GHz lags by 30 d behind that at 14.5 GHz, or τ_8−14.5 = 30 d. In comparison with the method of discrete correlation function (DCF), the time delay obtained by taking advantage of the TDCF method is much more rational for the multi-band correlation analysis of the 7 blazars.     

基于实测数据的综合脉冲星时研究

介绍了脉冲星计时在几个重要方面的应用,给出了脉冲星时间尺度的简要定义.基于澳大利亚PARKES天文台的毫秒脉冲星实测计时数据,建立了基于4颗毫秒脉冲星的综合脉冲星时,并将其稳定度σz与原子时进行比较,最后分析了影响综合脉冲星时的几个重要因素及其对脉冲星计时精度的影响与应用.     

A Method for locating Singularities in the Complex Time Domain for Integrable Dynamical Systems

A simple method for the determination of the position of singularities in the complex time domain for dynamical systems which are described by ordinary differential equations is presented. The method is designed for integrable separable systems whose solutions are not expressible in closed form. A direct consequence of this method is that it ‘closes’ the phase space. Simple physical meaning is given to the singularity position. This revised version was published online in July 2006 with corrections to the Cover Date.     

基于NPSVM模型的超新星识别方法

巡天观测与高能物理、黑洞天文等领域均有密切的联系.基于星系-超新星二分类问题,研究光谱数据预处理,结合余弦相似度改善PCA(Principal Component Analysis)光谱分解特征提取方法,用SDSS(the Sloan Digital Sky Survey)、WISeREP(the Weizmann Interactive Supernova data REPository)组成的5620条光谱数据集训练支持向量机,可以得到0.498%泛化误差的识别模型和新样本分类概率.使用Neyman-Pearson决策方法建立NPSVM(Neyman-Pearson Support Vector Machine)模型可进一步降低超新星的漏判率.     

Comparison between Windowed FFT and Hilbert-Huang Transform for Analyzing Time Series with Poissonian Fluctuations: A Case Study

Hilbert-Huang Transform (HHT) is a novel data analysis technique for nonlinear and non-stationary data. We present a time-frequency analysis of both simulated light curves and an X-ray burst from the X-ray burster 4U 1702-429 with both the HHT and the Windowed Fast Fourier Transform (WFFT) methods. Our results show that the HHT method has failed in all cases for light curves with Poissonian fluctuations which are typical for all photon counting instruments used in astronomy, whereas the WFFT method can sensitively detect the periodic signals in the presence of Poissonian fluctuations; the only drawback of the WFFT method is that it cannot detect sharp frequency variations accurately.