基于CEEMD与GA-SVR的猪肉价格集成预测模型

为提高猪肉价格预测的准确性,结合互补集合经验模态分解(CEEMD)的分解能力和基于遗传算法的支持向量回归(GA-SVR)的自适应预测功能,构建猪肉价格集成预测模型.首先为解决猪肉价格的复杂波动特征,通过CEEMD对猪肉价格分解得到本征模态函数(IMF)序列集;然后使用排序熵(PE)对IMF序列进行复杂度分析,进一步使用快速傅里叶变换方法(FFT)分解复杂度高的序列;再利用灰色关联度(GCD)对IMF序列集进行关联性分析,聚合相似IMF序列;最后基于各IMF序列的数据特征构建相应的GA-SVR预测模型,并将子序列的预测结果集成获得最终价格预测值.以中国集贸市场的猪肉价格为研究对象,实证结果表明,该集成预测模型在预测精度和方向性指标上,显著优于其他单预测模型和分解集成预测模型.     

非线性多尺度建模及其应用

针对非线性时间序列数据建模问题,首先对数据进行经验模态分解(Empirical Mode Decomposition,EMD),得到一系列反映序列演化过程的本征模态函数(Intrinsic mode function,IMF),然后对IMF进行Hilbert变换,利用各分量的Hilbert变换的振幅和频率建立非线性多尺度数学模型.以黄河某水质检测站溶解氧含量为例,应用EMD分解和Hilbert变换的结果,建立两个多尺度的数学模型.经过检验,模型的拟合精度、显著性和残差均符合要求.模型模拟结果中提取溶解氧多尺度信息,并且应用该模型进行预报,预报的相对误差小在0.08%~12.8%之间.     

EMD-SVM在南京市月平均气温预测中的应用

南京市月平均气温具有非平稳性、噪声大、序列宽频等特征.为了提高温预测精度,本文提出一种经验模态分解(EMD)和支持向量机(SVM)回归相组合的预测模型(EMD-SVM).首先应用EMD分解算法把南京市月平均气温分解成不同尺度的基本模态分量(IMF),再运用支持向量机回归模型对每个IMF预测,最后将预测结果重构得到南京市月平均气温预测值.结果表明:EMD-SVM模型预测与单一支持向量机回归模型预测相比,平均预测精度提高0.59度,是一种有效的预测气温的模型.     

基于VMD-ELM的农产品期货价格分解集成预测模型

为了捕捉农产品市场期货价格波动的复杂特征,进一步提高其预测精度,基于分解集成的思想,构建包含变分模态分解(VMD)和极限学习机(ELM)的分解集成预测模型。首先,利用VMD分解的自适应性和非递归性,选择VMD将复杂时间序列分解成多个模态分量(IMF)。其次,针对VMD分解关键参数模态数K的选取难题,提出基于最小模糊熵准则寻找最优K值的方法,有效避免模态混淆和端点效应问题,从而提升VMD的分解能力。最后,利用ELM强大的学习能力和泛化能力,对VMD分解得到的不同尺度子序列进行预测,集成得到最终预测结果。以CBOT交易所稻谷、小麦、豆粕期货价格作为研究对象,实证结果表明,该分解集成预测模型在预测精度和方向性指标上,显著优于单预测模型和其它分解集成预测模型,为农产品期货价格预测提供了一种新途径。     

混合区间多尺度分解的区间时间序列组合预测

鉴于传统预测方法一直基于“点”来衡量时间序列数据,然而现实生活中在给定的时间段内许多变量是有区间限制的,点值预测会损失波动性信息。因此,本文提出了一种基于混合区间多尺度分解的组合预测方法。首先,建立区间离散小波分解方法(IDWT)、区间经验模态分解方法(IEMD)和区间奇异普分析方法(ISSA)。其次,用本文构建的IDWT、IEMD和ISSA对区间时间序列进行多尺度分解,从而得到区间趋势序列和残差序列。然后,用霍尔特指数平滑方法(Holt's)、支持向量回归(SVR)和BP神经网络对区间趋势序列和残差序列进行组合预测得到三种分解方法下的区间时间序列预测值。最后,用BP神经网络对各预测结果进行集成得到区间时间序列最终预测值。同时,为证明模型的有效性进行了AQI空气质量的实证预测分析,结果表明,本文所提出基于混合区间多尺度分解的组合预测方法具有较高的预测精度和良好的适用性。     

基于SSA的金融时间序列自适应分解预测

提出了分解预测的思想,通过SSA将序列分解成低频与高频两部分,分别采用最小均方(LMS)自适应自回归移动平均(ARIMA)与LMS自适应自回归(AR)模型进行预测,然后将两者叠加便可得原始序列预测值.同时,为了更好地捕捉序列局部突变,缩减预测延迟,提高预测精度,对EaLMS算法(基于误差调整的LMS算法)参数进行修正并...     

一种基于局部回归的多尺度碳市场价格预测模型研究

碳市场价格呈现非线性、非平稳的复杂特性,准确预测具有较大的挑战。基于“分而治之”的思想,提出了一种基于局部回归的多尺度碳市场价格预测模型。提出的模型利用集成经验模态分解(EEMD)对碳市场价格时间序列进行分解。启发于EEMD局部特征分解的特点,对分解后的分量采用局部回归方法进行预测,然后将分量预测结果进行集成。采用的局部回归方法包括局部线性回归(LLP)、局部多项式回归、局部岭回归、局部主成分回归、局部偏最小二乘回归和局部套索回归。实验结果表明基于局部回归的多尺度预测模型具有优异的预测性能。在提出的模型中,EEMD-LLP结构简单且性能更为突出,进一步对EEMD-LLP参数的适应性进行探讨。与新近提出模型的对比结果表明了EEMD-LLP在碳市场价格预测中的有效性。     

非结构性数据驱动的混合分解集成碳交易价格组合预测

碳交易价格的有效预测有助于投资者合理决策以及政府制定科学的碳交易政策。本文提出一种非结构性数据驱动的混合分解集成碳交易价格组合预测方法。首先，基于百度指数获得碳交易相关非结构性数据，并利用主成分分析(PCA)方法提取其主成分。其次，对主成分序列与碳交易价格历史数据进行经验模态分解(EMD)、变分模态分解(VMD)与小波分解(WT),按频率高低重构后得到它们的高、低频序列和趋势项。然后，自适应选取自回归移动平均模型(ARIMA)、Holt指数平滑法和人工神经网络模型(ANN),结合非结构信息对碳价格的高、低频序列和趋势项进行预测。最后，基于BP神经网络等对三种分解方法的预测值分层集成，得到碳价格最终预测结果。对比实验结果显示，上述组合预测方法充分利用了多源信息，预测精度高且适用性良好。     

基于EEMD的LS-SVM和BP神经网络混合短期负荷预测

提出了基于总体平均经验模态分解(EEMD)、最小二乘支持向量机(LSSVM)和BP神经网络的实用综合短期负荷预测方法,进行电力系统短期负荷预测.首先运用EEMD方法将非平稳的负荷序列分解,然后根据分解后各分量的特点选用最佳的核函数,利用最小二乘支持向量机分别对各分量进行预测,最后对各分量预测结果采用BP神经网络重构得到最终的预测结果.对实测数据的分析表明基于该综合方法的电力系统短期负荷预测具有较高的精度.     

基于EMD方法的混沌信号中周期分量的提取

提出一种从Duffing振子产生的混沌信号中提取谐波分量的方法．依据任何信号由不同的固有简单振动模态组成的概念,利用经验模式分解（EMD）方法,将混沌信号分离为不同的内在模态函数(IMF),并在特定参数下从中分解出单一频率成分的谐波信号,从而成功地将混沌信号和谐波分量分离．仿真实验都表明该方法非常有效．     

Short-term prediction of wind power using EMD and chaotic theory

Due to the strong non-linear, complexity and non-stationary characteristics of wind farm power, a hybrid prediction model with empirical mode decomposition (EMD), chaotic theory, and grey theory is constructed. The EMD is used to decompose the wind farm power into several intrinsic mode function (IMF) components and one residual component. The grey forecasting model is used to predict the residual component. For the IMF components, identify their characteristics, if it is chaotic time series use largest Lyapunov exponent prediction method to predict. If not, use grey forecasting model to predict. Prediction results of residual component and all IMF components are aggregated to produce the ultimate predicted result for wind farm power. The ultimate predicted result shows that the proposed method has good prediction accuracy, can be used for short-term prediction of wind farm power.     

Transformation of chaotic nonlinear polynomial difference systems through Newton iterations

Chaotic sequences generated by nonlinear difference systems or ‘maps’ where the defining nonlinearities are polynomials, have been examined from the point of view of the sequential points seeking zeroes of an unknown functionf following the rule of Newton iterations. Following such nonlinear transformation rule, alternative sequences have been constructed showing monotonie convergence. Evidently, these are maps of the original sequences. For second degree systems, another kind of possibly less chaotic sequences have been constructed by essentially the same method. Finally, it is shown that the original chaotic system can be decomposed into a fast monotonically convergent part and a principal oscillatory part showing sharp oscillations. The methods are exemplified by the well-known logistic map, delayed-logistic map and the Hénon map.     

一种改进的混沌序列产生方法

根据灰度图像的二维直方图的特点,在已有的二维Arnold混沌系统的基础上,结合Bernstein形式的Bézier曲线的生成算法,给出了一种基于生成Bézier曲线的de Casteljau算法构造伪随机序列的方法,实验结果表明生成的二维序列不仅具有伪随机性,而且还具有在近似圆盘中随机分布的性质,这使得该伪随机序列更适合对灰度图像的二维灰度直方图进行基于混沌优化的图像分割.在此基础上,给出了一种基于混沌优化的二维最大熵的灰度图像分割算法,该算法对于含噪图像取得了良好的分割效果.     

Adaptive fuzzy observer based synchronization design and secure communications of chaotic systems

This paper proposes a synchronization design scheme based on an alternative indirect adaptive fuzzy observer and its application to secure communication of chaotic systems. It is assumed that their states are unmeasurable and their parameters are unknown. Chaotic systems and the structure of the fuzzy observer are represented by the Takagi–Sugeno fuzzy model. Using Lyapunov stability theory, an adaptive law is derived to estimate the unknown parameters and the stability of the proposed system is guaranteed. Through this process, the asymptotic synchronization of chaotic systems is achieved. The proposed observer is applied to secure communications of chaotic systems and some numerical simulation results show the validity of theoretical derivations and the performance of the proposed observer.     

基于三类模型的四大银行股票收益率预测研究

运用时间序列分析的预测方法,对四大银行的股票日对数收益率序列进行拟合与预测分析,分别构建ARMA模型、GARCH模型以及ARMA-GARCH组合模型,通过模型比较,实证分析表明:在拟合效果上,ARMA-GARCH模型的拟合优度优于ARMA模型和GARCH模型;在预测效果上,ARMA模型的预测效果最优,ARMA-GARCH模型次之.     

基于EM算法的混合自回归滑动平均模型的参数估计

研究了一类用于时间序列建模的混合自回归滑动平均模型.该模型是由m个ARMA分量经过混合得到的,给出了混合自回归滑动平均模型参数估计的期望极大化(EM)算法,从而得到了混合系数和分量模型的参数,通过仿真说明了其有效性.     

Strictly stationary solutions of multivariate ARMA equations with i.i.d. noise

We obtain necessary and sufficient conditions for the existence of strictly stationary solutions of multivariate ARMA equations with independent and identically distributed driving noise. For general ARMA(p, q) equations these conditions are expressed in terms of the coefficient polynomials of the defining equations and moments of the driving noise sequence, while for p =?1 an additional characterization is obtained in terms of the Jordan canonical decomposition of the autoregressive matrix, the moving average coefficient matrices and the noise sequence. No a priori assumptions are made on either the driving noise sequence or the coefficient matrices.     

An asymmetric image cryptosystem based on the adaptive synchronization of an uncertain unified chaotic system and a cellular neural network

Chaotic dynamics provide a fast and simple means to create an excellent image cryptosystem, because it is extremely sensitive to initial conditions and system parameters, pseudorandomness, and non-periodicity. However, most chaos-based image encryption schemes are symmetric cryptographic techniques, which have been proven to be more vulnerable, compared to an asymmetric cryptosystem. This paper develops an asymmetric image cryptosystem, based on the adaptive synchronization of two different chaotic systems, namely a unified chaotic system and a cellular neural network. An adaptive controller with parameter update laws is formulated, using the Lyapunov stability theory, to asymptotically synchronize the two chaotic systems. The synchronization controller is embedded in the image cryptosystem and generates a pair of asymmetric keys, for image encryption and decryption. Using numerical simulations, three sets of experiments are conducted to evaluate the feasibility and reliability of the proposed chaos-based image cryptosystem.     

Performance evaluation of CPPM modulation in multi-path environments

Chaotic pulse position modulation (CPPM) is a novel technique to communicate with chaotic signals based upon pulse trains in which the intervals between two pulses are determined by chaotic dynamics of a pulse generator. Using numerical simulations we show that CPPM offers excellent multi-path performance. We simulated the CPPM radio system, which is designed for a WLAN application and operates in the 2.4 GHz ISM frequency band with IEEE 802.11 compliant channel spacing. In this case, the average performance loss due the multi-path for CPPM is less than 5 dB.