INTERNET TRAFFIC DATA FLOW FORECAST BY RBF NEURAL NETWORK BASED ON PHASE SPACE RECONSTRUCTION

Characteristics of the Internet traffic data flow are studied based on the chaos theory. A phase space that is isometric with the network dynamic system is reconstructed by using the single variable time series of a network flow. Some parameters, such as the correlative dimension and the Lyapunov exponent are calculated, and the chaos characteristic is proved to exist in Internet traffic data flows. A neural network model is construct- ed based on radial basis function （RBF） to forecast actual Internet traffic data flow. Simulation results show that, compared with other forecasts of the forward-feedback neural network, the forecast of the RBF neural network based on the chaos theory has faster learning capacity and higher forecasting accuracy.

Study on resource quantity of surface water based on phase space reconstruction and neural network

Proposed a new method to disclose the complicated non-linearity structure of the water-resource system, introducing chaos theory into the hydrology and water resources field, and combined with the chaos theory and artificial neural networks. Training data construction and networks structure were determined by the phase space reconstruction, and establishing nonlinear relationship of phase points with neural networks, the forecasting model of the resource quantity of the surface water was brought forward. The keystone of the way and the detailed arithmetic of the network training were given. The example shows that the model has highly forecasting precision.

Neural network forecasting model based on phase space re-construction in water yield of mine

The neutral network forecasting model based on the phase space reconstruction was proposed. First, through reconstructing the phase space, the time series of single variable was done excursion and expanded into multi- dimension series which included the ergodic information and more rich information could be excavated. Then, on the basis of the embedding dimension of the time series, the structure form of neutral network was constructed, of which the node number in input layer was the embedding dimension of the time series minus 1, and the node number in output layers was 1. Finally, as an example, the model was applied for water yield of mine forecasting. The result shows that the model has good fitting accuracy and forecasting precision.

Deep learning approach to detect seizure using reconstructed phase space images

Epilepsy is a chronic neurological disorder that affects the function of the brain in people of all ages.It manifests in the electroencephalogram(EEG) signal which records the electrical activity of the brain.Various image processing,signal processing,and machine-learning based techniques are employed to analyze epilepsy,using spatial and temporal features.The nervous system that generates the EEG signal is considered nonlinear and the EEG signals exhibit chaotic behavior.In order to capture these nonlinear dynamics,we use reconstructed phase space(RPS) representation of the signal.Earlier studies have primarily addressed seizure detection as a binary classification(normal vs.ictal) problem and rarely as a ternary class(normal vs.interictal vs.ictal)problem.We employ transfer learning on a pre-trained deep neural network model and retrain it using RPS images of the EEG signal.The classification accuracy of the model for the binary classes is(98.5±1.5)% and(95±2)% for the ternary classes.The performance of the convolution neural network(CNN) model is better than the other existing statistical approach for all performance indicators such as accuracy,sensitivity,and specificity.The result of the proposed approach shows the prospect of employing RPS images with CNN for predicting epileptic seizures.

Prediction of elevator traffic flow based on SVM and phase space reconstruction

To make elevator group control system better follow the change of elevator traffic flow (ETF) in order to adjust the control strategy,the prediction method of support vector machine (SVM) in combination with phase space reconstruction has been proposed for ETF.Firstly,the phase space reconstruction for elevator traffic flow time series (ETFTS) is processed.Secondly,the small data set method is applied to calculate the largest Lyapunov exponent to judge the chaotic property of ETF.Then prediction model of ETFTS based on SVM is founded.Finally,the method is applied to predict the time series for the incoming and outgoing passenger flow respectively using ETF data collected in some building.Meanwhile,it is compared with RBF neural network model.Simulation results show that the trend of factual traffic flow is better followed by predictive traffic flow.SVM algorithm has much better prediction performance.The fitting and prediction of ETF with better effect are realized.

A Phase Space Reconstruction Based Approach to Throughput Prediction in Semiconductor Wafer Fabrication System

In order to manage and control semiconductor wafer fabrication system (SWFS) more effectively,the daily throughput prediction data of wafer fab are often used in the planning and scheduling of SWFS.In this paper,an artificial neural network (ANN) prediction method based on phase space reconstruction (PSR) and ant colony optimization (ACO) is presented,in which the phase space reconstruction theory is used to reconstruct the daily throughput time series,the ANN is used to construct the daily throughput prediction model,and the ACO is used to train the connection weight and bias values of the neural network prediction model.Testing with factory operation data and comparing with the traditional method show that the proposed methodology is effective.

PARAMETERS DETERMINATION METHOD OF PHASE-SPACE RECONSTRUCTION BASED ON DIFFERENTIAL ENTROPY RATIO AND RBF NEURAL NETWORK

Phase space reconstruction is the first step of recognizing the chaotic time series.On the basis of differential entropy ratio method,the embedding dimension opt m and time delay t are optimal for the state space reconstruction could be determined.But they are not the optimal parameters accepted for prediction.This study proposes an improved method based on the differential entropy ratio and Radial Basis Function(RBF)neural network to estimate the embedding dimension m and the time delay t,which have both optimal characteristics of the state space reconstruction and the prediction.Simulating experiments of Lorenz system and Doffing system show that the original phase space could be reconstructed from the time series effectively,and both the prediction accuracy and prediction length are improved greatly.

Intelligent Space All-Optical Network Technology

Microwave transmission in a space network is greatly restricted due to precious radio spectrum resources. To meet the demand for large-bandwidth, global seamless coverage and on-demanding access, the Space All-Optical Network(SAON) becomes a promising paradigm. In this paper, the related space optical communications and network programs around the world are first briefly introduced. Then the intelligent Space All-Optical Network(i-SAON), which can be deemed as an advanced SAON, is illustrated, with the emphasis on its features of high survivability, sensing and reconfiguration intelligence, and large capacity for all optical load and switching. Moreover, some key technologies for i-SAON are described, including the rapid adjustment and control of the laser beam direction, the deep learning-based multi-path anti-fault routing, the intelligent multi-fault diagnosis and switching selection mechanism, and the artificial intelligence-based spectrum sensing and situational forecasting.

Identification of Atrial Fibrillation Using Complex Network Similarity

We investigate the use of complex network similarity for the identification of atrial fibrillation. The similarity of the network is estimated via the joint recurrence plot and Hamming distance. Firstly, we transform multi-electrodes epicardium signals recorded from dogs into the recurrence complex network. Then, we extract features representing its similarity. Finally, epicardium signals are classified utilizing the classification and regression tree with extracted features. The method is validated using 1000 samples including 500 atrial fibrillation cases and 500 normal sinus ones. The sensitivity, specificity and accuracy of the identification are 98.2%, 98.8% and 98.5% respectively. This experiment indicates that our approach may lay a foundation for the prediction of the onset of atrial fibrillation.

Effects of Model Structural Complexity and Data Pre-Processing on Artificial Neural Network (ANN) Forecast Performance for Hydrological Process Modelling

The choice of a particular Artificial Neural Network (ANN) structure is a seemingly difficult task;worthy of relevance is that there is no systematic way for establishing a suitable architecture. In view of this, the study looked at the effects of ANN structural complexity and data pre-processing regime on its forecast performance. To address this aim, two ANN structural configurations: 1) Single-hidden layer, and 2) Double-hidden layer feed-forward back propagation network were employed. Results obtained revealed generally that: a) ANN comprised of double hidden layers tends to be less robust and converges with less accuracy than its single-hidden layer counterpart under identical situations;b) for a univariate time series, phase-space reconstruction using embedding dimension which is based on dynamical systems theory is an effective way for determining the appropriate number of ANN input neurons, and c) data pre-processing via the scaling approach excessively limits the output range of the transfer function. In specific terms considering extreme flow prediction capability on the basis of effective correlation: Percent maximum and minimum correlation coefficient (Rmax% and Rmin%), on the average for one-day ahead forecast during the training and validation phases respectively for the adopted network structures: 8 7 5 (i.e., 8 input nodes, 7 nodes in the hidden layer, and 5 output nodes in the output layer), 8 5 2 5 (8 nodes in the input layer, 5 nodes in the first hidden layer, 2 nodes in the second hidden layer, and 5 nodes in the output layer), and 8 4 3 5 (8 nodes in the input layer, 4 nodes in the first hidden layer, 3 nodes in the second hidden layer, and 5 nodes in the output layer) gave: 101.2, 99.4;100.2, 218.3;93.7, 95.0 in all instances irrespective of the training algorithm (i.e., pooled). On the other hand, in terms of percent of correct event prediction, the respective performances of the models for both low and high flows during the training and validation phases, respectively were: 0.78, 0.96: 0.65, 0.87;0.76, 0.93: 0.61, 0.83;and 0.79, 0.96: 0.65, 0.87. Thus, it suffices to note that on the basis of coherence or regularity of prediction consistency, the ANN model: 8 4 3 5 performed better. This implies that though the adoption of large hidden layers vis-à-vis corresponding large neuronal signatures could be counter-productive because of network over-fitting, however, it may provide additional representational power. Based on the findings, it is imperative to note that ANN model is by no means a substitute for conceptual watershed modelling, therefore, exogenous variables should be incorporated in streamflow modelling and forecasting exercise because of their hydrologic evolutions.

相空间重构与改进SMA优化SVR的网络流量预测

为提高网络流量预测精度,提出结合相空间重构与改进黏菌优化支持向量回归的预测模型。为解决黏菌算法收敛慢、易得局部最优的不足,引入3种形态对立学习对种群进行初始化,提高种群多样性;利用非线性反馈因子更新机制,均衡全局搜索与局部开发;设计柯西-高斯混合变异对最优解变异,扩展搜索空间,避免陷入局部最优。利用改进黏菌算法对支持向量回归优化调参,有效解决超参初值敏感缺陷,提高学习精度和收敛速度,以此构建网络流量预测模型。实验结果表明,改进模型预测误差更小,能够实现高精度和实时性预测要求。

基于分解集成及不确定理论的碳价格预测

准确的碳市场价格预测是碳排放交易市场相关政策制定和碳金融发展的基础.为消除碳市场价格原始序列存在的非线性、非平稳性、高噪声性和不确定性,准确预测碳市场价格,论文将不确定理论、集合经验模态分解(ensemble empirical mode decomposition,简称EEMD)和径向基神经网络(radial basis function,简称RBF)相结合,构建了碳市场价格预测模型,并将其应用于广东省碳市场价格预测.首先通过EEMD算法和fine-to-coarse方法对原始的碳市场价格数据进行分解和重构,得到具有不同变化规律的高频项和低频项,并将其代入RBF神经网络进行训练,然后采用不确定理论,对低频项的输出权重进行不确定性分析,对残差趋势项采用线性回归进行拟合,最后将3个子项的预测结果进行集成求和得到最终的碳市场价格预测值.实证结果表明无论是在均方根误差(root mean square error,简称RMSE)、平均绝对误差(mean absolute error,简称MAE)还是在平均绝对百分比误差(mean absolute percentage error,简称MAPE)指标方面,论文模型在碳市场价格预测方面都比其他预测模型更具优势,预测结果更准确.

一种适用三电平并联系统的虚拟矢量调制策略

直流侧中点电位平衡和零序环流抑制是保证三电平逆变器并联系统正常运行的重要问题,然而基于冗余小矢量对调节的环流抑制方案进一步增加空间矢量脉宽调制(space vector pulse width modulation,SVPWM)中点电流无法对称消除的场景,加剧双机系统中点电位不平衡程度。该文基于虚拟矢量合成思路,提出一种满足零序环流抑制和中点电位平衡需求的双机虚拟矢量调制方法。该方法重新定义虚拟中矢量,通过缩短中矢量作用时间和增加参与合成的对称冗余小矢量以减少中点电流不平衡场景;分析不同扇区不同控制目标下虚拟矢量合成规则,利用相占空比法优化虚拟矢量的作用顺序,减少矢量合成过程中的开关次数。与传统SVPWM策略下中点电压变化量的对比分析表明,所提方法能够削弱环流抑制和中点电位平衡的耦合关系。最后,利用Starsim平台实验验证该策略的有效性。

配电网间歇性重燃电弧模型的建立与断续弧光接地故障特征分析研究

电弧间歇性重燃是配电网单相接地故障最显著的特征。现有的电弧模型甚少考虑电弧间歇性重燃特性,导致无法精确描述断续弧光接地特征,进而影响继电保护动作。为此,论文提出一种间歇性重燃电弧模型的建立方法,并在此基础上对断续弧光接地故障特征进行了分析。弧道阻抗的随机变化是电弧间歇性重燃的重要标志,故论文重点围绕弧道阻抗变化的随机性和重燃时间间隔的随机性开展间歇性重燃电弧模型的研究。黑盒电弧模型中,Cassie-Mayr联合模型能完整的描述电弧电流从大电流到小电流的变化过程,但存在从大电流变化为小电流的判据模糊,转换过程突变的问题。为此,论文通过引入连续过渡函数解决上述问题。同时,为描述弧道电阻的变化特性,利用Fermi函数对联合模型中Mayr模型和Cassie模型进行权重分配。以改进的Cassie-Mayr单次燃弧模型为基础,根据工频熄弧理论,通过设置燃弧时间长短表征间歇性重燃的随机性,从而建立了间歇性重燃电弧模型。利用该模型,对典型10kV配电网单辐射型网架结构的接地故障进行模拟仿真,采用快速傅里叶变换(fast Fourier transform,FFT)和小波包分析提取了不同条件下故障电压、电流、高次谐波、零序分量以及频率分布等故障特征。研究结果表明:改进后Cassie-Mayr联合模型不但解决了电弧电流从大电流到小电流的转换突变问题,且不同模型权重占比的分配更能准确地表征实际燃弧弧道阻抗变化的随机性;通过设置电弧燃弧时间长短,准确地描述间歇性重燃的随机性;电弧断续时刻为非整数周期下的过电压、过电流幅值高于整数周期;电缆线路增大了故障线路电流,过电流可达3.81~7.20pu,不利于熄弧;大电流系统故障相零序电流主频在0~400Hz,小电流系统故障相零序电流主频在1200~1600Hz。

基于多变量相空间重构和径向基函数神经网络的综合能源系统电冷热超短期负荷预测

为解决能源危机问题,提高能源利用率,综合能源系统(integrated energy system,IES)成为发展创新型能源系统的重要方向。准确的多元负荷预测对IES的经济调度和优化运行有着重要的影响,而借助混沌理论能够进一步挖掘IES多元负荷潜在的耦合特性。提出了一种基于多变量相空间重构(multivariate phase space reconstruction,MPSR)和径向基函数神经网络(radial basis function neural network,RBFNN)相结合的IES超短期电冷热负荷预测模型。首先,分析了IES中能源子系统之间的耦合关系,运用Pearson相关性分析定量描述多元负荷和气象特征的相关性。然后,采用C-C法对时间序列进行MPSR以进一步挖掘电冷热负荷和气象特征在时间上的耦合特性。最后,利用RBFNN模型对电冷热负荷间耦合关系进行学习并预测。实验结果表明,所提方法有效挖掘并学习电冷热负荷在时间上的耦合特性,且在不同样本容量下具有良好且稳定的预测效果。

基于时序拓扑数据分析的电力电缆局部放电模式识别

在电力电缆局部放电(PD)模式识别时,相位图谱以及统计特征往往因区分度不足而影响识别精度。为此,提出了一种基于时序拓扑数据分析(TDA)的局放特征提取和识别方法。首先,提出一种符号熵和粒子群优化(PSO)相结合的重构参数选择方法,将预处理后的局放时域信号进行相空间重构,并生成三维局放数据点云;然后,基于TDA方法提取持续同调特征,据此生成持续散点图及持续条形码,计算并可视化表达为贝蒂曲线;最后,将贝蒂曲线输入1D-CNN模型,对4种典型局放缺陷模式进行识别并开展对比实验。实验结果表明,该方法对相空间重构时延参数的选取更加准确,且TDA特征具备良好的区分度,相比其他以相位图谱及统计特征为输入的模型,该方法整体识别准确率最高可提升15.34%,达到98.55%。

基于复杂网络的交通序列数据特性

为了进一步研究交通流特性,采用复杂网络方法对交通序列数据进行分析。提出了箱型图-聚类算法模型用于识别和填充初始数据中的缺失值和异常值;通过相空间重构方法将一维数据重构为网络节点,选取连接阈值确定网络节点的连接关系,将交通序列数据构建为复杂网络,对复杂网络的结构和定量指标进行分析。研究结果表明交通序列数据复杂网络的结构一定程度上可以反映路段的交通流状态。该结果有助于优化数据预处理方法,拓展复杂网络在交通序列数据研究中的应用。

基于MPSR和IRBM的电力系统中长期负荷预测

针对电力系统中长期负荷波动大及不确定因素导致负荷预测误差较大的问题,提出了一种基于多变量相空间重构(MPSR)和改进受限波尔兹曼机(IRBM)的电力系统中长期负荷预测方法。首先,利用多元线性回归分析方法分析天气因素与电负荷之间的相关性,并将其与电负荷序列组成多变量时间序列;然后,利用C-C法确定每一时间序列的最优嵌入维数和时间延迟,实现多变量相空间重构;最后,采用多变量相空间重构建立的数据集训练电力系统负荷预测模型,同时利用梯度优化法对参数进行优化,得到预测模型。结果表明:相比长短期记忆神经网络和粒子群优化BP神经网络,所提出的预测方法有较高的精准度。

A prediction comparison between univariate and multivariate chaotic time series

The methods to determine time delays and embedding dimensions in the phase space delay reconstruction of multivariate chaotic time series are proposed. Three nonlinear prediction methods of multivariate chaotic time series including local mean prediction, local linear prediction and BP neural networks prediction are considered. The simulation results obtained by the Lorenz system show that no matter what nonlinear prediction method is used, the prediction error of multivariate chaotic time series is much smaller than the prediction error of univariate time series, even if half of the data of univariate time series are used in multivariate time series. The results also verify that methods to determine the time delays and the embedding dimensions are correct from the view of minimizing the prediction error.

基于TNN-BL模型的低压配电网断零与缺相故障检测方法研究

低压配电网中因断零与缺相故障对电网公司造成的安全隐患和经济损失一直是电网公司迫切解决的难题,随着智能化检测设备在电网中普及,可利用智能电表采集的低压侧负载电压和各序电流数据开展故障检测。首先,建立基于Transformer神经网络(TNN)和双向长短期记忆(Bi-LSTM)的混合模型TNN-BL;其次,通过选择合适的损失函数和正则化函数完善模型以进一步提高模型检测性能;最后,采用南网数据集对模型性能进行试验验证。试验结果表明,该方法拥有更有效的特征提取能力,相比于其他故障检测方法具有更高的检测准确度和更强的鲁棒性。