Detection of low-dimensional chaos in buildings energy consumption time series

In this paper, nonlinear time series modeling techniques are applied to analyze building energy consumption data. The time series were obtained for the benchmark data set Proben 1, and comes from the first energy prediction contest, the Great Building Energy Predictor Shootout I, organized by ASHRAE. The phase space, which describes the evolution of the behavior of a nonlinear system, is reconstructed using the delay embedding theorem suggested by TAKENS. The embedding parameters, e.g. the delay time and the embedding dimension are estimated using the average mutual information (AMI) of the data and the false nearest neighbor (FNN) algorithm, respectively. Nonlinearity was detected, by applying the surrogate data sets method.Numerically estimated non-integral fractal dimension and a positive Lyapunov exponent are not necessarily sufficient indication of chaos; therefore we apply a more stringent criterion, developed by Gao and Zheng, which is based on the logarithmic displacement of time-dependent exponent curves, and show that these data are chaotic.Based on this analysis and proof, we then calculate the correlation dimension of the resulting attractor and the largest Lyapunov exponent. The correlation dimension 3.47 and largest Lyapunov exponent 0.047 are estimated. These results indicate that chaotic characteristics obviously exist in the specific energy consumption data set, and thus techniques based on phase space dynamics can be used to analyze and predict buildings energy use.     

Neural network method for determining embedding dimension of a time series

A method is described for determining the optimal short-term prediction time-delay embedding dimension for a scalar time series by training an artificial neural network on the data and then determining the sensitivity of the output of the network to each time lag averaged over the data set. As a byproduct, the method identifies any intermediate time lags that do not influence the dynamics, thus permitting a possible further reduction in the required embedding dimension. The method is tested on four sample data sets and compares favorably with more conventional methods including false nearest neighbors and the ‘plateau dimension’ determined by saturation of the estimated correlation dimension. The proposed method is especially advantageous when the data set is small or contaminated by noise. The trained network could be used for noise reduction, forecasting, and estimating the dynamical and geometrical properties of the system that produced the data, such as the Lyapunov exponent, entropy, and attractor dimension.     

Multivariate polynomial regression for identification of chaotic time series

Multivariate polynomial regression was used to generate polynomial iterators for time series exhibiting autocorrelations. A stepwise technique was used to add and remove polynomial terms to ensure the model contained only those terms that produce a statistically significant contribution to the fit. An approach is described in which datasets are divided into three subsets for identification, estimation, and validation. This produces a parsimonious global model that is can greatly reduce the tendency towards undesirable behaviours such as overfitting or instability. The technique was found to be able to identify the nonlinear dynamic behaviour of simulated time series, as reflected in the geometry of the attractor and calculation of multiple Lyapunov exponents, even in noisy systems.

The technique was applied to times series data obtained from simulations of the Lorenz and Mackey – Glass equations with and without measurement noise. The model was also used to determine the embedding dimension of the Mackey – Glass equation.     

基于遗传算法和半参数回归的神经网络集成股市预测研究

股票时间序列预测在经济和管理领域具有重要的应用前景,也是很多商业和金融机构成功的基础.首先利用奇异谱分析对股市时间序列重构,降低噪声并提取趋势序列.再利用C-C算法确定股市时间序列的嵌入维数和延迟阶数,对股市时间序列进行相空间重构,生成神经网络的学习矩阵.进一步利用Boosting技术和不同的神经网络模型,生成神经网络集成个体.最后采用带有惩罚项的半参数回归模型进行集成,并利用遗传算法选择最优的光滑参数,以此建立遗传算法和半参数回归的神经网络集成股市预测模型.通过上证指数开盘价进行实例分析,与传统的时间序列分析和其他集成方法对比,发现该方法能获得更准确的预测结果.计算结果表明该方法能充分反映股票价格时间序列趋势,为金融时间序列预测提供一个有效方法.     

Optimizing Markovian modeling of chaotic systems with recurrent neural networks

In this paper, we propose a methodology for optimizing the modeling of an one-dimensional chaotic time series with a Markov Chain. The model is extracted from a recurrent neural network trained for the attractor reconstructed from the data set. Each state of the obtained Markov Chain is a region of the reconstructed state space where the dynamics is approximated by a specific piecewise linear map, obtained from the network. The Markov Chain represents the dynamics of the time series in its statistical essence. An application to a time series resulted from Lorenz system is included.     

Attractors of reaction‐diffusion systems in unbounded domains and their spatial complexity

The nonlinear reaction‐diffusion system in an unbounded domain is studied. It is proven that, under some natural assumptions on the nonlinear term and on the diffusion matrix, this system possesses a global attractor ?? in the corresponding phase space. Since the dimension of the attractor happens to be infinite, we study its Kolmogorov's ?‐entropy. Upper and lower bounds of this entropy are obtained. Moreover, we give a more detailed study of the attractor for the spatially homogeneous RDE in ?ⁿ. In this case, a group of spatial shifts acts on the attractor. In order to study the spatial complexity of the attractor, we interpret this group as a dynamical system (with multidimensional “time” if n > 1) acting on a phase space ??. It is proven that the dynamical system thus obtained is chaotic and has infinite topological entropy. In order to clarify the nature of this chaos, we suggest a new model dynamical system that generalizes the symbolic dynamics to the case of the infinite entropy and construct the homeomorphic (and even Lipschitz‐continuous) embedding of this system into the spatial shifts on the attractor. Finally, we consider also the temporal evolution of the spatially chaotic structures in the attractor and prove that the spatial chaos is preserved under this evolution. © 2003 Wiley Periodicals, Inc.     

Global dynamics of diffusive Hindmarsh–Rose equations with memristors

Global dynamics of the diffusive Hindmarsh–Rose equations with memristors as a new proposed model for neuron dynamics are investigated in this paper. We prove the existence and regularity of a global attractor for the solution semiflow through uniform analytic estimates showing the higher-order dissipative property and the asymptotically compact characteristics of the solutions by the approach of Kolmogorov–Riesz theorem. The quantitative bounds of the regions containing this global attractor respectively in the state space and in the regular space are explicitly expressed by the model parameters.     

基于季节性RBF神经网络的月度市场需求预测研究

本文提出一种季节性神经网络预测模型,对具有季节性变化的产品月度市场需求进行预测.在Matlab语言环境下,用傅立叶周期分析法得到时间序列的周期长度;借鉴嵌入理论,提出了确定季节性神经网络输入维数的策略;利用计算机程序搜索,确定最优参数;通过合理插值,重构样本集.仿真实验表明,该模型的预测精度明显高于其他几个常用的季节预测模型.     

带有阻尼项的广义对称正则长波方程的指数吸引子

考虑了带有阻尼项的广义对称正则长波方程的整体快变动力学．证明了与该方程有关的非线性半群的挤压性质和指数吸引子的存在性．对指数吸引子的分形维数的上界也进行了估计．     

带有热记忆的非均匀柔性结构的长时间动力行为

本文研究了带有热效应的非均匀柔性结构方程，并且该热效应符合Coleman-Gurtin定律.利用半群方法，建立了系统的整体适定性.主要结论是该系统的长时间动力行为.本文证明了系统的拟稳定性，整体吸引子的存在性以及整体吸引子具有有限的分形维数.此外，还证明了指数吸引子的存在性.