Application of uncertainty reasoning based on cloud model in time series prediction

Time series prediction has been successfully used in several application areas, such as meteoro-logical forecasting, market prediction, network traffic forecasting, etc. , and a number of techniques have been developed for modeling and predicting time series. In the traditional exponential smoothing method, a fixed weight is assigned to data history, and the trend changes of time series are ignored. In this paper, an uncertainty reasoning method, based on cloud model, is employed in time series prediction, which uses cloud logic controller to adjust the smoothing coefficient of the simple exponential smoothing method dynamically to fit the current trend of the time series. The validity of this solution was proved by experiments on various data sets.     

Auto-Regressive Models of Non-Stationary Time Series with Finite Length

To analyze and simulate non-stationary time series with finite length, the statistical characteristics and auto-regressive (AR) models of non-stationary time series with finite length are discussed and studied. A new AR model called the time varying parameter AR model is proposed for solution of non-stationary time series with finite length. The auto-covariances of time series simulated by means of several AR model sare analyzed. The result shows that the new AR model can be used to simulate and generate a new time series with the auto-covariance same as the original time series. The size curves of cocoon filaments regarded as non-stationary time series with finite length are experimentally simulated. The simulation results are significantly better than those obtained so far, and illustrate the availability of the time varying parameter AR model. The results are useful for analyzing and simulating non-stationary time series with finite length.     

Regional adaptive delaminating algorithm for rapid prototyping

With the development of rapid prototyping (RP) technology, the contradiction between fabricating speed and precision becomes more and more acute. In order to solve the contradiction, a regional effective adaptive delaminating algorithm is proposed in this paper. It is different from the traditional method that the regional adaptive delaminating method divides slicing contours into several regions and adaptively delaminates in respective region only according to the contours information. Namely, this method can not only adaptively delaminate at different height of CAD model but also adaptively delaminate different regions in one slicing layer. Furthermore, because the two-dimensional contour data adopted in this method are necessary for rapid prototyping system, this regional effective adaptive delaminating method is suitable for all RP systems. A comparative study is used to analyze the effectiveness of this method, which can demonstrate that this method can increase the fabricating speed and reduce the running cost under the condition of ensuring fabricating precision.     

Dynamic prediction of building subsidence deformation with data-based mechanistic self-memory model

This paper describes a building subsidence deformation prediction model with the self-memorization principle.According to the non-linear specificity and monotonic growth characteristics of the time series of building subsidence deformation,a data-based mechanistic self-memory model considering randomness and dynamic features of building subsidence deformation is established based on the dynamic data retrieved method and the self-memorization equation.This model first deduces the differential equation of the building subsidence deformation system using the dynamic retrieved method,which treats the monitored time series data as particular solutions of the nonlinear dynamic system.Then,the differential equation is evolved into a difference-integral equation by the self-memory function to establish the self-memory model of dynamic system for predicting nonlinear building subsidence deformation.As the memory coefficients of the proposed model are calculated with historical data,which contain useful information for the prediction and overcome the shortcomings of the average prediction,the model can predict extreme values of a system and provide higher fitting precision and prediction accuracy than deterministic or random statistical prediction methods.The model was applied to subsidence deformation prediction of a building in Xi’an.It was shown that the model is valid and feasible in predicting building subsidence deformation with good accuracy.     

Trajectory tracking control of quantum systems

For quantum state trajectory tracking of density matrix in Liouville equation of quantum systems,with the help of concept in quantum system control,one can apply unitary transformation both to controlled system and free-evolutionary target system such as to change the time-variant and non-stationary target system into a stationary state.Therefore,the quantum state trajectory tracking problem becomes a steering one.State steering control law of the system transformed is designed by means of the Lyapunov stability theorem.Finally,numerical simulation experiments are given for a five-level energy quantum system.The comparison analysis of original system’s trajectory tracking with other method illustrates the advantage in control time of the method proposed.     

Fashion Color Forecasting by Applying an Improved Back Propagation Neural Network

Fashion color forecasting is one of the most important factors for fashion marketing and manufacturing. Several models have been applied by previous researchers to conduct fashion color forecasting. However, few convincing forecasting systems have been established. A prediction model for fashion color forecasting was established by applying an improved back propagation neural network (BPNN) model in this paper. Successive six-year fashion color palettes, released by INTERCOLOR, were used as learning information for the neural network to develop a reliable prediction model. Colors in the palettes were quantified by PANTONE color system. Additionally, performance of the established model was compared with other GM(1, 1) models. Results show that the improved BPNN model is suitable to predict future fashion color trend.     

Bearing capacity analysis of reinforced concrete L-shaped cross-section under biaxial eccentric compression

The ultimate capacity of short L-shaped reinforced concrete (RC) members subjected to biaxial eccentric compression is investigated to provide a new tool for structure engineers. On the basis of nonlinear analysis of RC arbitrary cross-section, a new simplified cross-section numerical integration method is suggested. The cross-section is divided into several triangles, and each one is integrated with Gauss integration method. And L-shaped cross-sections subjected to biaxial eccentric compression are analyzed. The interaction curves of (N-M) and (My-Mz) for L-shaped sections are derived. The analytical results are well consistent with the experimental data and other analytical results.     

Development of a Chinese land data assimilation system:its progress and prospects

The objective of land data assimilation is to merge multi-source observations into the dynamics of land surface model for improving the estimation of land surface states.We have developed a land data assimilation system for China's land territory.In this sys- tem,the Common Land Model and Simple Biosphere Model 2 are used to simulate land surface processes.The radiative transfer models of thawed and frozen soil,snow,lake,and vegetation are used as observation operators to transfer model predictions into estimated bright- ness temperatures.A Monte-Carlo based sequential filter,the ensemble Kalman filter,is implemented as data assimilation method to inte- grate modeling and observation.The system is capable of assimilating passive microwave remotely sensed data such as special sensor mi- crowave/imager (SSM/I),TRMM microwave imager (TMI),and advanced microwave scanning radiometer enhanced for EOS (AMSR- E) and the conventional in situ measurements of soil and snow.A spatiotemporally consistent assimilated dataset for soil moisture,soil temperature,snow and frozen soil,with a spatial resolution of 0.25 degree and temporal resolution of one hour,has been produced.This paper introduces the development of Chinese land data assimilation system and the progress made on data assimilation algorithms,land sur- face modeling,microwave remote sensing of land surface hydrological variables,and the preparation of atmospheric forcing data.The dis- tinct characteristics and challenges of developing land data assimilation system and the perspectives for future development are also dis- cussed.     

Weighted LBF for spaceborne general bistatic SAR processing

Loffeld＇s bistatic formula （LBF） is the first two-dimensional analytic point target reference spectrum derived for general bistatic SAR frequency domain focusing. The phase history is expanded in Taylor series around the individual points of stationary phase of the trans- mitter-target and target-receiver phase histories, respectively, and thus the common bistatic stationary phase point can be obtained using the method of stationary phase. Unfortunately, it shows limitations for extreme bistatic configurations, namely the highly squinted mode and space-surface application. The weighted LBF （WLBF） is proposed in this paper based on the different contributions of total phase modulation from the transmitter and receiver. The formulae we derived are compared with that of the original literature. The extreme bistatic stripmap SAR data can be focused using WLBF, which accommodates the spaceborne squint geometry using the modified effective velocity solution. A point target simulation example is presented to verify the accuracy of the new WLBF spectrum.     

Seasonal and inter-annual variations of length of day and polar motion observed by SLR in 1993―2006

A time series of length of the day (LOD) and polar motion (PM) were estimated from the range data measured by the satellite laser ranging technique (SLR) to LAGEOS 1/2 through 1993 to 2006. Compared with EOPC04 released by the International Earth Rotation and Reference Systems Service (IERS), the root mean squares errors for LOD, X and Y of PM are 0.0067 milliseconds (ms), 0.18 milli-arc-sec- onds (mas) and 0.20 mas, respectively. Then the time series are analyzed with the wavelet transformation and least squares method. Wavelet analysis shows that there are the obvious seasonal and inter- annual variations of LOD and PM, but the annual variation cannot be distinguished from the Chandler variation because these two frequencies are very close. The trends and periodic variations of LOD and PM are given in the least squares sense. LOD changes with the annual and semiannual periods. The annual and Chandler variations for PM are also detected, but the semiannual motion for PM is not found. The trend rate of the LOD change in 1993―2006 is ?0.18 ms per year, and the difference from the well- known 1.7 ms per century showed that the trend rate is diverse in different periods possibly. The trend rates of PM in the X and Y directions are 2.25 and 1.67 mas per year respectively, and the North Pole moves to 36.5°E relative to the crust, which is different from the direction of Greenland.     

基于小波分解的径流非线性预测

利用小波变换原理将具有非平稳特征的径流序列进行分解，使其平稳项与随机项分离。对平稳项采用传统的AR模型加以预测，而通过对随机项的混沌特征研究，发现其具有明显的混沌特征，进而提出了基于非线性混沌动力学的预测模型方法。最后通过小波对所提出的AR NCDF预测模型预测结果予以重构，实现对原始径流序列的预测。该方法通过实例验证具有较高的精度，是一种实际可行的方法。     

基于小波分解与自回归树的时间序列预测新方法

小波分解可以将非平稳时间序列分解成多层近似意义上的平稳时间序列,本文首先采用小波分解将非平稳时间序列分解,在分解后的各层时间序列上构造自回归树模型,采用贝叶斯方法学习决策树的结构与变量,并对分解后的时间序列进行预测,最后采用小波重构方法将分解后的各层时间序列重构,得到原始时间序列的预测值。以2007年海关统计的重点出口商品量的数据为例,对中国出口贸易的走势进行分析和预测。结果表明,本文的方法比传统时间序列预测方法精度高,可以很好地应用于非平稳时间序列的预测。     

基于ARMA模型的振动信号建模与预测

振动信号是反映系统故障的重要信息,预测振动信号的趋势是系统状态监测与故障诊断中的一个重要内容。本文提出了一种采用时间序列模型来分析和预测非线性随机振动信号的方法,建立了描述振动信号的数学模型。首先将现场测得的非平稳振动信号转化成标准正态平稳时间序列,然后利用这些参考信号建立时间序列模型,并采用非线性最小二乘法进行模型参数估计,最后利用该模型对信号进行预测。应用结果表明该模型能够很好地拟合振动信号时间序列,并取得了一定的预测精度,可以达到预测要求。     

基于小波分解的设备状态预测方法

首次提出将小波分解应用于非平稳时间序列的预测中,通过小波分解将非平稳时间序列分解为多层近似意义上的平稳时间序列,并且用ＡＲ（ｎ）模型对分解后的时间序列进行预测,进而得到最终的预测结果．将该方法应用于压缩机轴承座磨损的趋势预测中,通过与基于ＢＰ网络的预测方法相比较表明：该方法预测精度高,而且预测速度快,可以有效地应用设备状态的预测和设备故障趋势的分析中．     

灰色预测和时序预测的探讨

本文讨论了灰色模型,特别是GM(1,1)模型的特点和适用范围,并将GM(1,1)模型和时序AR(n)模型结合起来(称为组合模型),对我国轻工业产量发展指数等三个项目分别进行了组合模型预测。结果表明,在一般GM模型中引入AR模型可显著提高预测的准确度;在非平稳时序建模中引入GM模型,可作为提取趋势项的另一种方法。文中还从预测的角度将灰色模型和时序模型进行了比较和分析,对“灰”的物理概念进行了初步探讨。     

基于ARIMA模型的全国电视收视率实证分析

现代电视媒体经营中,要想在日益完善的经济竞争机制中增加收益,对于电视媒体来说至关重要,本文通过对全国样本城市的收视率数据进行了周一至周日的走势分析,并在走势相同的情况下对周四的收视率进行了时间序列建模分析,在将原始非平稳序列经二阶差分后进行了ARIMA建模。     

交通流的时间序列建模及预测

采用时间序列ARMA模型进行交通流量趋势预测，建模时利用现场测得的交通流非平稳时间序列进行差分变换及标准化处理，从而转化成标准正态平稳时间序列；模型参数估计采用极大似然估计法，并根据AIC准则为模型定阶；最后通过实测数据进行验证．结果表明该ARMA模型能够较好地拟合交通流时间序列并可获得较高的中短期预测精度，因而可用于动态交通信号控制．     

一种非线性非平稳时间序列预测建模方法

提出了一种基于经验模式分解和支持向量回归的非线性、非平稳时间序列预测建模方法.首先,针对时间序列的非平稳特征,通过经验模式分解将其分解为若干个本征模式分量,使其中每个分量均成为平稳序列;其次,对每个本征模式分量,基于支持向量回归建立相应的平稳时间序列预测模型;最后,再一次利用支持向量回归对这些预测模型进行非线性组合,得到非线性、非平稳时间序列的预测模型.仿真实验和工程应用均表明,所提的预测建模方法与传统的基于支持向量回归的建模方法相比,具有较高的精度,说明该方法对于非线性、非平稳时间序列的预测是有效的.     

产品需求量非平稳时序的ANN-ARMA预测模型

针对基于非平稳时序的产品需求量预测方法存在的问题,研究了人工神经网络(ANN)与自回归滑动平均(ARMA)模型的集成建模与预测方法. 产品需求量的非平稳时序可分解为确定项和随机项两个部分,用人工神经网络模型拟合确定项,以表示非平稳的变化趋势;用自回归滑动平均模型拟合随机项,以表示平稳的随机成分. 将两个模型的预测值之和作为产品需求量的优化预测值. 仿真结果表明,集成模型的预测精度高于单一的人工神经网络模型.     

一种基于小波分析理论的灰色预测方法

提出一种基于小波分析理论的灰色预测方法．该方法通过小波分解将非平稳时间序列分解到多个尺度上以减少序列的随机性，然后建立灰色预测模型对分解后的时间序列分别进行预测，从而得到原始时间序列的预测值．并通过对上证指数的预测，结果表明该方法预测效果良好，优于一般灰色预测方法．