结合可视图的多状态交通流时间序列特性分析

交通流时间序列的研究主要采用数据挖掘和机器学习的方法,这些"黑箱"挖掘方法很难直观反映序列特性.为增强交通流时间序列及其特征分析的可视化性,结合可视图理论来构建交通流时间序列的关联网络,从复杂网络角度实现交通流时间序列的特性分析.在网络构建的过程中,考虑到不同交通状态下交通流表征具有的差异性,首先利用交通流参量的相关性对交通流状态进行分类,然后构建不同交通状态下的时间序列复杂网络,并对这些网络的特征属性给出统计分析,如度分布、聚类系数、网络直径、模块化等.研究表明,可视图法可为交通流时间序列映射到网络提供有效途径,并且不同状态下交通流时间序列构建的复杂网络的模块化、聚类系数和度分布等统计特征呈现一定的变化规律,为交通流运行态势的研究提供了可视化的分析角度.     

一种新型广义RBF神经网络在混沌时间序列预测中的研究

提出了一种新颖的广义径向基函数神经网络模型,其径向基函数(RBF)的形式由生成函数确定.然后,给出了易实现的梯度学习算法,同时为了进一步提高网络的收敛速度和网络性能,又给出了基于卡尔曼滤波的动态学习算法.为了验证网络的学习性能，采用基于卡尔曼滤波算法的新型广义RBF网络预测模型对Mackey-Glass混沌时间序列和Henon映射进行了仿真.结果表明,所提出的新型广义RBF神经网络模型能快速、精确地预测混沌时间序列,是研究复杂非线性动力系统辨识和控制的一种有效方法. 关键词： 广义径向基函数神经网络 卡尔曼滤波 梯度下降学习算法 混沌时间序列 预测     

基于时间序列符号化模式表征的有向加权复杂网络

时间序列复杂网络分析近些年已发展成为非线性信号分析领域的一个国际热点课题.为了能更有效地挖掘时间序列(特别是非线性时间序列)中的结构特征,同时简化时间序列分析的复杂度,提出了一种新的基于时间序列符号化结合滑窗技术模式表征的有向加权复杂网络建网方法.该方法首先按照等概率区段划分的方式将时间序列做符号化处理,结合滑窗技术确定不同时刻的符号化模式作为网络的节点;然后将待分析时间序列符号化模式的转换频次和方向作为网络连边的权重和方向,从而建立时间序列有向加权复杂网络.通过对Logistic系统不同参数设置对应的时间序列复杂网络建网测试结果表明,相比经典的可视图建网方法,本文方法的网络拓扑能更简洁、直观地展示时间序列的结构特征.进而,将本文方法应用于规则排列采集的自然风场信号分析,其网络特性指标能较准确地预测采集信号的排布规律,而可视图建网方法的网络特性指标没有任何规律性的结果.     

Visitor flow pattern of Expo 2010

Expo 2010 Shanghai China was a successful, splendid, and unforgettable event, leaving us with valuable experi- ences. The visitor flow pattern of the Expo is investigated in this paper. The Hurst exponent, the mean value, and the standard deviation of visitor volume indicate that the visitor flow is fractal with long-term stability and correlation as well as obvious fluctuation in a short period. Then the time series of visitor volume is converted into a complex network by using the visibility algorithm. It can be inferred from the topological properties of the visibility graph that the network is scale-free, small-world, and hierarchically constructed, confirming that the time series are fractal and a close relationship exists among the visitor volumes on different days. Furthermore, it is inevitable that will be some extreme visitor volumes in the original visitor flow, and these extreme points may appear in a group to a great extent. All these properties are closely related to the feature of the complex network. Finally, the revised linear regression is performed to forecast the next-day visitor volume based on the previous 10-day data.     

Relaxation in cooperative systems: Use of mixture virial coefficients

The coefficients in power series in the variable time that describe relaxation in a cooperative system can be calculated using a combinatorial approach where one considers how many ways one can introduce a given number of properly defined events in a system. The coefficients obtained in this manner can be related to the equilibrium virial coefficients for a mixture. If one assumes rapid internal equilibration, the relaxation process can be expressed completely in terms of the viral coefficients for a mixture with at most one solute particle, or, in some cases, just the virial coefficients for a single-component system. Thus, equilibrium virial coefficients can give useful information about the time evolution of processes in cooperative systems.     

Chaotic time series prediction using least squares support vector machines

We propose a new technique of using the least squares support vector machines (LS-SVMs) for making one-step and multi-step prediction of chaotic time series. The LS-SVM achieves higher generalization performance than traditional neural networks and provides an accurate chaotic time series prediction. Unlike neural networks‘ training that requires nonlinear optimization with the danger of getting stuck into local minima, training LS-SVM is equivalent to solving a set of linear equations. Thus it has fast convergence. The simulation results show that LS-SVM has much better potential in the field of chaotic time series prediction.     

Characterizing and modelling cyclic behaviour in non-stationary time series through multi-resolution analysis

A method based on wavelet transform is developed to characterize variations at multiple scales in non-stationary time series. We consider two different financial time series, S&P CNX Nifty closing index of the National Stock Exchange (India) and Dow Jones industrial average closing values. These time series are chosen since they are known to comprise of stochastic fluctuations as well as cyclic variations at different scales. The wavelet transform isolates cyclic variations at higher scales when random fluctuations are averaged out; this corroborates correlated behaviour observed earlier in financial time series through random matrix studies. Analysis is carried out through Haar, Daubechies-4 and continuous Morlet wavelets for studying the character of fluctuations at different scales and show that cyclic variations emerge at intermediate time scales. It is found that Daubechies family of wavelets can be effectively used to capture cyclic variations since these are local in nature. To get an insight into the occurrence of cyclic variations, we then proceed to model these wavelet coefficients using genetic programming (GP) approach and using the standard embedding technique in the reconstructed phase space. It is found that the standard methods (GP as well as artificial neural networks) fail to model these variations because of poor convergence. A novel interpolation approach is developed that overcomes this difficulty. The dynamical model equations have, primarily, linear terms with additive Padé-type terms. It is seen that the emergence of cyclic variations is due to an interplay of a few important terms in the model. Very interestingly GP model captures smooth variations as well as bursty behaviour quite nicely.      

基于模糊树模型的混沌时间序列预测

应用模糊树模型,对混沌时间序列进行建模和预测.该方法可以根据建模数据在空间中的分布信息,基于二叉树结构自适应划分输入空间,得到模糊子空间,在与叶节点对应的子空间上建立线性函数作为模糊规则的后件,用隶属度函数将各分片线性函数光滑连接,最后得到一个精度比较高的非线性映射.通过对Mackey-Glass、Lorenz和Henon混沌时间序列的建模和预测研究,仿真结果表明,该方法具有建模精度高、运行速度快、泛化能力强、预测步数多、适用范围广等优点. 关键词： 模糊树模型 混沌时间序列 预测     

Ordinal Pattern Dependence in the Context of Long-Range Dependence

Ordinal pattern dependence is a multivariate dependence measure based on the co-movement of two time series. In strong connection to ordinal time series analysis, the ordinal information is taken into account to derive robust results on the dependence between the two processes. This article deals with ordinal pattern dependence for a long-range dependent time series including mixed cases of short- and long-range dependence. We investigate the limit distributions for estimators of ordinal pattern dependence. In doing so, we point out the differences that arise for the underlying time series having different dependence structures. Depending on these assumptions, central and non-central limit theorems are proven. The limit distributions for the latter ones can be included in the class of multivariate Rosenblatt processes. Finally, a simulation study is provided to illustrate our theoretical findings.     

短期风速时间序列混沌特性分析及预测

针对短期风速时间序列的预测问题进行了研究. 首先通过0-1混沌测试法确定短期风速时间序列具有混沌特性. 采用相空间重构技术, 利用C-C算法确定延迟时间, G-P 算法确定嵌入维数. 然后提出一种参数在线修正的最小二乘支持向量机预测模型, 采用改进的粒子群算法进行预测模型中参数的优化. 最后通过仿真对比实验表明提出的预测方法在预测精度、预测误差、预测效果方面都要优于其他常见的预测方法, 证明该预测方法是有效的.     

Pattern interdependent network of cross-correlation in multivariate time series

Cross-correlation of a bivariate time series induces interdependencies between local patterns in the two series, which cooperatively exhibit in turn the structure of the cross-correlation. However, this structure is lost in the procedure of statistical average in time series analysis. In this paper a new concept called pattern interdependent network is proposed to display the structure of cross-correlation, in which the nodes are unique local patterns and the linkages are co-occurring frequencies of the unique local patterns in the series. The performance is illustrated by the bivariate series generated with the Gaussian process and the auto-regressive fractionally integrated moving average (ARFIMA) model. It is found that the cross-correlation and the scaling behaviors dominate the pattern of backbone structure (the set of the nodes and the set of linkages) and the symmetry of the network, respectively. The ARFIMA model can reproduce the structural behaviors of cross-correlations in U.S. stock markets. This concept provides us with a new method for detecting the structure of couplings between time series in various fields, such as clinical pathological signals.     

基于时间序列的航天器遥测数据预测算法

在航天器遥测数据预测领域,基于时间序列的预测方法有着广阔的应用前景;时间序列有一明显的特性就是记忆性,记忆性是指时间数列中的任一观测值的表现皆受到过去观测值影响;它的基本思想是根据观测数据的特点为数据建立尽可能合理的统计模型,利用模型的统计特性解释数据的统计规律,以期达到预报的目的;提出了采用模式识别和参数估计的方法,结合航天器遥测动态数据,建立关于航天器遥测数据的时序预测模型,对航天器遥测数据趋势进行检测和预报。     

Adaptive step-size modified fractional least mean square algorithm for chaotic time series prediction

This paper presents an adaptive step-size modified fractional least mean square （AMFLMS） algorithm to deal with a nonlinear time series prediction. Here we incorporate adaptive gain parameters in the weight adaptation equation of the original MFLMS algorithm and also introduce a mechanism to adjust the order of the fractional derivative adaptively through a gradient-based approach. This approach permits an interesting achievement towards the performance of the filter in terms of handling nonlinear problems and it achieves less computational burden by avoiding the manual selection of adjustable parameters. We call this new algorithm the AMFLMS algorithm. The predictive performance for the nonlinear chaotic Mackey Glass and Lorenz time series was observed and evaluated using the classical LMS, Kernel LMS, MFLMS, and the AMFLMS filters. The simulation results for the Mackey glass time series, both without and with noise, confirm an improvement in terms of mean square error for the proposed algorithm. Its performance is also validated through the prediction of complex Lorenz series.     

Pattern synchrony in electrical signals related to earthquake activity

We apply the cross sample entropy method to geoelectrical time series collected from independent channels (North-South and East-West directions) monitored at two sites located in Mexico, to assess the presence of pattern synchrony between the signals, particularly in the proximity of earthquakes. To our best knowledge, this method has not been applied yet for the study of electrical signals related to earthquake activity. Moreover, we introduce the multiscale pattern synchrony analysis by extending the multiscale entropy technique to calculate the cross-entropy between two signals, which represents a novel approach to the study of pattern synchrony. The results obtained suggest that in the vicinity of an earthquake the geoelectrical signals exhibit pattern synchrony that persists for long sequences and through multiple scales, in addition to the presence of correlations in each channel.     

基于平均场支持向量机的混沌时间序列预测

提出了一种新的基于支持向量机的混沌时间序列预测方法，该方法利用平均场理论使支持向量机的学习过程变得简单高效。同时由于该方法将支持向量机的参数近似为高斯分布的，因此采用平均场理论能够容易的求解这些参数，这样获得的支持向量机的参数比传统的基于二次规划的算法更加精确，而且学习速度更快。最后利用该方法对嵌入维数与模型的泛化能力关系进行了探讨，并利用Mackey-Glass时间序列对该方法进行了验证，结果表明：该预测方法能精确地预测混沌时间序列,而且在混沌时间序列的嵌入维数未知时也能取得比较好的预测效果.这一结论预示着平均场支持向量机是一种研究混沌时间序列的有效方法.     

Prediction of chaotic time series based on modified minimax probability machine regression

Long-term prediction of chaotic time series is very difficult,for the Chaos restricts predictability.in this paper a new method is studied to model and predict chaotic time series based on minimax probability machine regression （MPMR）. Since the positive global Lyapunov exponents lead the errors to increase exponentially in modelling the chaotic time series, a weighted term is introduced to compensate a cost function. Using mean square error （MSE） and absolute error （AE） as a criterion, simulation results show that the proposed method is more effective and accurate for multistep prediction. It can identify the system characteristics quite well and provide a new way to make long-term predictions of the chaotic time series.[第一段]     

MULTISTAGE ADAPTIVE HIGHER-ORDER NONLINEAR FINITE IMPULSE RESPONSE FILTERS FOR CHAOTIC TIME SERIES PREDICTIONS

A multistage adaptive higher-order nonlinear finite impulse response (MAHONFIR) filter is proposed to predict chaotic time series. Using this approach, we may readily derive the decoupled parallel algorithm for the adaptation of the coefficients of the MAHONFIR filter, to guarantee a more rapid convergence of the adaptive weights to their optimal values. Numerical simulation results show that the MAHONFIR filters proposed here illustrate a very good performance for making an adaptive prediction of chaotic time series.     

一种预测混沌时间序列的模糊神经网络方法

给出了一种预测混沌时间序列的模糊神经网络及其学习方法，给出的方法能直接从数据中提取模糊规则，经过优化得到最佳模糊规则库，并利用神经网络的自学习功能修改隶属函数的参数和网络的权值，减少了规则的匹配过程，加快了推理速度，增强了网络的自适应能力. 使用该神经网络及其学习方法对Lorenz混沌时间序列进行了预测仿真研究，试验结果表明给出的预测工具和方法是有效的. 关键词： 模糊神经网络 模糊规则提取 混沌时间序列预测     

基于数据融合的多变量相空间重构方法

针对单变量时间序列和多变量时间序列相空间重构所存在的问题,提出一种新的多变量融合的相空间重构方法. 通过Bayes估计理论,将多变量在同一相空间中进行相点的最优融合,得到了更为理想的融合相空间. 应用所提出的方法对Lorenz系统及耦合Rssler系统进行了多变量融合的相空间重构. 通过多变量重构图与单变量重构图的比较,发现基于数据融合的多变量相空间重构图包含了所有单变量相空间重构图的重要信息,使重构的相空间更加完备,较全面地反映出吸引子的全貌信息. 最后应用该方法对转子油膜涡动故障得到的多变量时间序列 关键词： 多变量时间序列 相空间重构 数据融合 Bayes估计     

具有选择与遗忘机制的极端学习机在时间序列预测中的应用

针对训练样本贯序输入时的极端学习机 (ELM)训练问题,提出一种具有选择与遗忘机制的极端学习机 (SF-ELM),并研究了其在混沌时间序列预测中的应用. SF-ELM以逐次增加新训练样本的方式实现在线训练,通过引入遗忘因子以减弱旧训练样本的影响,同时以泛化能力为判断依据,对其输出权值进行选择性递推更新. 混沌时间序列在线预测实例表明,SF-ELM是一种有效的ELM在线训练模式. 相比于在线贯序极端学习机,SF-ELM具有更快的在线训练速度和更高的在线预测精度,因此更适于混沌时间序列在线预测. 关键词： 混沌时间序列 时间序列预测 神经网络 极端学习机