The application of Hölder exponent to traffic congestion warning

In this paper, we applied multifractal modeling techniques to analyze the traffic data collected from the Beijing Yuquanying. The results indicated that multifractal characteristics obviously exist in the traffic system; the degree of fractality of these traffic data tends to increase as the traffic system becomes congested; the Hölder exponent that measures the local rate of fractality may be used as indicators to predict the presence of the traffic congestion.     

Factorizing multivariate time series operators

Motivated by problems occurring in the empirical identification and modelling of a n-dimensional ARMA time series X(t) we study the possibility of obtaining a factorization (I + a₁B + … + a_pB^p) X(t) = [Π_i=1^p (I ? α_iB)] X(t), where B is the backward shift operator. Using a result in [3] we conclude that as in the univariate case such a factorization always exists, but unlike the univariate case in general the factorization is not unique for given a₁, a₂,…, a_p. In fact the number of possibilities is limited upwards by $\text{(np)!}\text{(n!)}{}^{\mathrm{p}}$, there being cases, however, where this maximum is not reached. Implications for the existence and possible use of transformations which removes nonstationarity (or almost nonstationarity) of X(t) are mentioned.     

Markov inequalities on partially ordered spaces

Let X(ω) be a random element taking values in a linear space $\text{X}$ endowed with the partial order ≤; let $\text{G}$₀ be the class of nonnegative order-preserving functions on $\text{X}$ such that, for each g∈$\text{G}$₀, E[g(X)] is defined; and let $\text{G}$₁?$\text{G}$₀ be the subclass of concave functions. A version of Markov's inequality for such spaces in P(X ≥ x) ≤ inf_$\text{G}$0E[g(X)]/g(x). Moreover, if E(X) = ξ is defined and if Jensen's inequality applies, we have a further inequality P(X≥x) ≤ inf_$\text{G}$1E[g(X)]/g(x) ≤ inf_$\text{G}$1g(ξ)/g(x). Applications are given using a variety or orderings of interest in statistics and applied probability.     

过滤窗最小二乘支持向量机的混沌时间序列预测

传统的滑动窗策略只是简单且机械地将最远的数据移出窗口, 而将最近的数据移进窗口. 针对这种遗忘策略存在的缺陷, 提出了过滤窗策略. 过滤窗采用"优胜劣汰"的选择机制, 将对模型贡献比较大的数据留在窗口当中. 将过滤窗和最小二乘支持向量回归机相结合, 提出了过滤窗最小二乘支持向量回归机. 与滑动窗最小二乘支持向量回归机相比较, 过滤窗最小二乘支持向量回归机具有较小的计算量, 需要较短的窗口长度就能达到和滑动窗最小二乘支持向量回归机几乎相同的预测精度, 而较短的窗口长度又预示着较少的计算量和较好的实时性. 混沌时间序列在线建模和预测的实例表明了过滤窗最小二乘支持向量回归机的有效性和可行性. 关键词： 混沌时间序列 支持向量机 滑动窗 过滤窗     

Perturbation theory in a framework of iteration methods

In a previous paper [J. Phys. A 36 (2003) 11807], we introduced the ‘asymptotic iteration method’ for solving second-order homogeneous linear differential equations. In this Letter, we study perturbed problems in quantum mechanics and we use the method to find the coefficients in the perturbation series for the eigenvalues and eigenfunctions directly, without first solving the unperturbed problem.     

Adomian Decomposition Method Using Integrating Factor

This paper proposes a new Adomian decomposition method by using integrating factor.Nonlinear models are solved by this method to get more reliable and efficient numerical results.It can also solve ordinary differential equations where the traditional one fails.Besides,the complete error analysis for this method is presented.     

基于protel DXP 2004的RLC电路仿真

用protel DXP 2004软件对RLC电路的暂态过程进行仿真实验分析,主要介绍了仿真电路的创建,并结合电路理论仿真了电路在欠阻尼、临界阻尼和过阻尼三种阻尼状态下的暂态过程。     

基于SARIMA-BP神经网络组合方法的MODIS叶面积指数时间序列建模与预测

植被叶面积指数(LAI)时间序列的建模及预测是陆面过程模型和遥感数据同化方法的重要组成部分。MODIS数据产品MOD15A2是目前应用最为广泛的LAI数据源之一,然而MODIS LAI时间序列产品包含了一些低质量的数据,例如由于云层、气溶胶等的影响,该产品在时间和空间上缺乏连续性。MODIS LAI时间序列包含线性部分和外在干扰产生的非线性部分,单一的线性方法或非线性方法都不能对其精确建模和预测。首先利用Savitzky-Golay(SG)滤波和线性插值平滑受到干扰的LAI时间序列,然后采用季节自回归积分滑动平均(SARIMA)方法、BP神经网络方法及二者的组合方法(SARIMA-BP)对MODIS LAI时间序列进行建模及预测。在SARIMA-BP神经网络组合方法中,各自在线性与非线性建模的优势得以充分发挥,其中SARIMA方法用于建模及预测LAI时间序列中的线性部分,BP神经网络方法用于对非线性残差部分进行建模及预测。实验结果显示：SG滤波和线性插值后的LAI时间序列比原LAI时间序列更平滑;SARIMA-BP神经网络组合方法的决定系数为0.981,比SARIMA和BP神经网络的0.941和0.884更接近于1;SARIMA-BP神经网络组合方法的预测值同观测值之间的相关系数为0.991,高于SARIMA(0.971)和BP神经网络(0.942)的相关系数。由此得出结论：SARIMA-BP神经网络组合方法对MODIS LAI时间序列具有更好的适应性,其建模和预测准确性高于SARIMA方法或BP神经网络方法。     

Herd formation and information transmission in a population: non-universal behaviour

We present generalized dynamical models describing the sharing of information, and the corresponding herd behavior, in a population based on the recent model proposed by Eguıluz and Zimmermann (EZ) [Phys. Rev. Lett. 85, 5659 (2000)]. The EZ model, which is a dynamical version of the herd formation model of Cont and Bouchaud (CB), gives a reasonable model for the formation of clusters of agents and for actions taken by clusters of agents. Both the EZ and CB models give a cluster size distribution characterized by a power law with an exponent -5/2. By introducing a size-dependent probability for dissociation of a cluster of agents, we show that the exponent characterizing the cluster size distribution becomes model-dependent and non-universal, with an exponential cutoff for large cluster sizes. The actions taken by the clusters of agents generate the price returns, the distribution of which is also characterized by a model-dependent exponent. When a size-dependent transaction rate is introduced instead of a size-dependent dissociation rate, it is found that the distribution of price returns is characterized by a model-dependent exponent while the exponent for the cluster-size distribution remains unchanged. The resulting systems provide simplified models of a financial market and yield power law behaviour with an easily tunable exponent. Received 31 December 2001     

绝缘栅双极晶体管固体开关技术研究

 采用二只1.2kV,75A的绝缘栅双极晶体管（IGBT）串联，对用于加速器的IGBT固体开关技术进行了研究；通过采取动态电压和静态电压平衡措施来解决IGBT串联中的瞬态电压平衡问题。在不同工作条件下，串联IGBT开关工作性能稳定，重复性好。在纯阻性负载上得到了上升时间约300ns、下降时间约164μs、峰值电压1.8kV的输出。