基于RBF-AR的船舶变形极短期预报

统一空间基准是海上作战平台实现精准探测打击的重要保证,而船体角变形的存在将严重影响空间基准的建立。针对这一问题,提出一种基于状态相依自回归（state-dependent auto-regressive, SD-AR）与径向基（radial basis function, RBF）神经网络的极短期变形预报方法,实现船体角形变的实时预报,为后续角变形的补偿提供依据。不同于传统的时间序列预报方法,该模型用一组RBF网络来逼近SD-AR模型中的函数系数,并采用一种结构化的非线性参数优化方法（structured nonlinear parameter optimization method, SNPOM）辨识该模型。基于该RBF-AR预报模型,给出了船舶变形预报算法设计并进行了仿真实验。实验结果表明,该方法在船体变形预测精度上优于传统时间序列预测方法,具有较好的应用前景。     

Learning the structure of dynamic Bayesian networks from time series and steady state measurements

Dynamic Bayesian networks (DBN) are a class of graphical models that has become a standard tool for modeling various stochastic time-varying phenomena. In many applications, the primary goal is to infer the network structure from measurement data. Several efficient learning methods have been introduced for the inference of DBNs from time series measurements. Sometimes, however, it is either impossible or impractical to collect time series data, in which case, a common practice is to model the non-time series observations using static Bayesian networks (BN). Such an approach is obviously sub-optimal if the goal is to gain insight into the underlying dynamical model. Here, we introduce Bayesian methods for the inference of DBNs from steady state measurements. We also consider learning the structure of DBNs from a combination of time series and steady state measurements. We introduce two different methods: one that is based on an approximation and another one that provides exact computation. Simulation results demonstrate that dynamic network structures can be learned to an extent from steady state measurements alone and that inference from a combination of steady state and time series data has the potential to improve learning performance relative to the inference from time series data alone.     

Improved estimation of electricity demand function by using of artificial neural network,principal component analysis and data envelopment analysis

Due to various seasonal and monthly changes in electricity consumption and difficulties in modeling it with the conventional methods, a novel algorithm is proposed in this paper. This study presents an approach that uses Artificial Neural Network (ANN), Principal Component Analysis (PCA), Data Envelopment Analysis (DEA) and ANOVA methods to estimate and predict electricity demand for seasonal and monthly changes in electricity consumption. Pre-processing and post-processing techniques in the data mining field are used in the present study. We analyze the impact of the data pre-processing and post-processing on the ANN performance and a 680 ANN-MLP is constructed for this purpose. DEA is used to compare the constructed ANN models as well as ANN learning algorithm performance. The average, minimum, maximum and standard deviation of mean absolute percentage error (MAPE) of each constructed ANN are used as the DEA inputs. The DEA helps the user to use an appropriate ANN model as an acceptable forecasting tool. In the other words, various error calculation methods are used to find a robust ANN learning algorithm. Moreover, PCA is used as an input selection method, and a preferred time series model is chosen from the linear (ARIMA) and nonlinear models. After selecting the preferred ARIMA model, the Mcleod–Li test is applied to determine the nonlinearity condition. Once the nonlinearity condition is satisfied, the preferred nonlinear model is selected and compared with the preferred ARIMA model, and the best time series model is selected. Then, a new algorithm is developed for the time series estimation; in each case an ANN or conventional time series model is selected for the estimation and prediction. To show the applicability and superiority of the proposed ANN-PCA-DEA-ANOVA algorithm, the data regarding the Iranian electricity consumption from April 1992 to February 2004 are used. The results show that the proposed algorithm provides an accurate solution for the problem of estimating electricity consumption.     

Structured parameter optimization method for the radial basis function-based state-dependent autoregressive model

An off-line structured nonlinear parameter optimization method (SNPOM) for accelerating the computational convergence of parameter estimation of the radial basis function-based state-dependent autoregressive (RBF-AR) model is proposed. Using the method, all the parameters of the RBF-AR model may be optimized automatically and simultaneously. The proposed method combines the advantages of the Levenberg-Marquardt algorithm in nonlinear parameter optimization and the least-squares method in linear parameter estimation. Case studies on two complex time series and a nonlinear chemical reaction process show that the proposed parameter optimization method exhibits significantly accelerated convergence when compared with the classic version of the Levenberg-Marquardt algorithm, and to some hybrid algorithms such as the evolutionary programming algorithm.     

Chaotic time series method combined with particle swarm optimization and trend adjustment for electricity demand forecasting

Electricity demand forecasting plays an important role in electric power systems planning. In this paper, nonlinear time series modeling technique is applied to analyze electricity demand. Firstly, the phase space, which describes the evolution of the behavior of a nonlinear system, is reconstructed using the delay embedding theorem. Secondly, the largest Lyapunov exponent forecasting method (LLEF) is employed to make a prediction of the chaotic time series. In order to overcome the limitation of LLEF, a weighted largest Lyapunov exponent forecasting method (WLLEF) is proposed to improve the prediction accuracy. The particle swarm optimization algorithm (PSO) is used to determine the optimal weight parameters of WLLEF. The trend adjustment technique is used to take into account the seasonal effects in the data set for improving the forecasting precision of WLLEF. A simulation is performed using a data set that was collected from the grid of New South Wales, Australia during May 14–18, 2007. The results show that chaotic characteristics obviously exist in electricity demand series and the proposed prediction model can effectively predict the electricity demand. The mean absolute relative error of the new prediction model is 2.48%, which is lower than the forecasting errors of existing methods.     

A non-homogeneous dynamic Bayesian network with a hidden Markov model dependency structure among the temporal data points

In the topical field of systems biology there is considerable interest in learning regulatory networks, and various probabilistic machine learning methods have been proposed to this end. Popular approaches include non-homogeneous dynamic Bayesian networks (DBNs), which can be employed to model time-varying regulatory processes. Almost all non-homogeneous DBNs that have been proposed in the literature follow the same paradigm and relax the homogeneity assumption by complementing the standard homogeneous DBN with a multiple changepoint process. Each time series segment defined by two demarcating changepoints is associated with separate interactions, and in this way the regulatory relationships are allowed to vary over time. However, the configuration space of the data segmentations (allocations) that can be obtained by changepoints is restricted. A complementary paradigm is to combine DBNs with mixture models, which allow for free allocations of the data points to mixture components. But this extension of the configuration space comes with the disadvantage that the temporal order of the data points can no longer be taken into account. In this paper I present a novel non-homogeneous DBN model, which can be seen as a consensus between the free allocation mixture DBN model and the changepoint-segmented DBN model. The key idea is to assume that the underlying allocation of the temporal data points follows a Hidden Markov model (HMM). The novel HMM–DBN model takes the temporal structure of the time series into account without putting a restriction onto the configuration space of the data point allocations. I define the novel HMM–DBN model and the competing models such that the regulatory network structure is kept fixed among components, while the network interaction parameters are allowed to vary, and I show how the novel HMM–DBN model can be inferred with Markov Chain Monte Carlo (MCMC) simulations. For the new HMM–DBN model I also present two new pairs of MCMC moves, which can be incorporated into the recently proposed allocation sampler for mixture models to improve convergence of the MCMC simulations. In an extensive comparative evaluation study I systematically compare the performance of the proposed HMM–DBN model with the performances of the competing DBN models in a reverse engineering context, where the objective is to learn the structure of a network from temporal network data.     

A moving-average filter based hybrid ARIMA–ANN model for forecasting time series data

A suitable combination of linear and nonlinear models provides a more accurate prediction model than an individual linear or nonlinear model for forecasting time series data originating from various applications. The linear autoregressive integrated moving average (ARIMA) and nonlinear artificial neural network (ANN) models are explored in this paper to devise a new hybrid ARIMA–ANN model for the prediction of time series data. Many of the hybrid ARIMA–ANN models which exist in the literature apply an ARIMA model to given time series data, consider the error between the original and the ARIMA-predicted data as a nonlinear component, and model it using an ANN in different ways. Though these models give predictions with higher accuracy than the individual models, there is scope for further improvement in the accuracy if the nature of the given time series is taken into account before applying the models. In the work described in this paper, the nature of volatility was explored using a moving-average filter, and then an ARIMA and an ANN model were suitably applied. Using a simulated data set and experimental data sets such as sunspot data, electricity price data, and stock market data, the proposed hybrid ARIMA–ANN model was applied along with individual ARIMA and ANN models and some existing hybrid ARIMA–ANN models. The results obtained from all of these data sets show that for both one-step-ahead and multistep-ahead forecasts, the proposed hybrid model has higher prediction accuracy.     

Nonlinear controller construction based on a model with state-dependent representation for a nonlinear system

This paper addresses a method of nonlinear controller construction based on a model with a state-dependent representation for a nonlinear system. In this method, the controller construction and generation of manipulated values are separated. A nonlinear system and its controller are firstly expressed by the coefficients of the state-dependent representation without any approximation. At the stage of controller implementation for the nonlinear system, the manipulated values are calculated by means of an algorithm of the numerical analysis. The properties of the proposed method are analysed. With the analytical considerations and simulation studies, the proposed method is compared with several nonlinear control methods, such as exact linearization method and the linear approximation method, and its merits are verified.     

基于动态贝叶斯网模型的股指收益率序列预测

证券市场预测,是当前研究的热点和难点.动态贝叶斯网(DBNs),能够学习变量间的概率依存关系及其随时间变化的规律,表达时间序列蕴含的潜在信息.利用DBNs方法,在证券心理分析技术的基础上.建立中国证券指数的日收益率预测模型.文中使用上海证券交易所综合指数日收益率数据对模型进行训练与预测.在离散量预测环境下,模型能达到80.12%预测命中率,在采用混合高斯(GMM)分布的连续量预测中,模型的平均绝对比例误差(MAPE)指标<1%.低于BP神经网络和GARCH-BP神经网络,而且累计误差增长稳定.说明:在市场高噪声的情况下,模型具有良好的稳定性和预测能力.     

Adaptive learning of dynamic Bayesian networks with changing structures by detecting geometric structures of time series

A dynamic Bayesian network (DBN) is one of popular approaches for relational knowledge discovery such as modeling relations or dependencies, which change over time, between variables of a dynamic system. In this paper, we propose an adaptive learning method (autoDBN) to learn DBNs with changing structures from multivariate time series. In autoDBN, segmentation of time series is achieved first through detecting geometric structures transformed from time series, and then model regions are found from the segmentation by designed finding strategies; in each found model region, a DBN model is established by existing structure learning methods; finally, model revisiting is developed to refine model regions and improve DBN models. These techniques provide a special mechanism to find accurate model regions and discover a sequence of DBNs with changing structures, which are adaptive to changing relations between multivariate time series. Experimental results on simulated and real time series show that autoDBN is very effective in finding accurate/reasonable model regions and gives lower error rates, outperforming the switching linear dynamic system method and moving window method.

基于深度孪生自回归网络的无监督异常用电检测

异常用电检测旨在识别出不符合正常用电规律或者违反用电合约的用电行为。针对现有基于重构的检测方法依赖标记的正常样本和难以捕捉复杂时间依赖性的问题,提出一种基于深度孪生自回归网络的无监督异常用电行为检测模型（DSAD）。所提模型通过两个孪生自回归子网络来分别独立地对无标记的输入数据进行重构,再将两个子网络的重构误差相结合来预测数据中的正常样本,并利用多头自注意力机制来有效地捕捉时间依赖性、周期性和随机性等复杂特征。在大规模时序数据集和国家电网真实用电数据集上进行实验所获得的结果表明,所提模型在AUC以及AP等性能指标上取得了更好的检测效果。     

Nonlinear system identification using two-dimensional wavelet-based state-dependent parameter models

This article presents a nonlinear system identification approach that uses a two-dimensional (2-D) wavelet-based state-dependent parameter (SDP) model. In this method, differing from our previous approach, the SDP is a function with respect to two different state variables, which is realised by the use of a 2-D wavelet series expansion. Here, an optimised model structure selection is accomplished using a PRESS-based procedure in conjunction with orthogonal decomposition (OD) to avoid any ill-conditioning problems associated with the parameter estimation. Two simulation examples are provided to demonstrate the merits of the proposed approach.     

Stability analysis of RBF network-based state-dependent autoregressive model for nonlinear time series

Varying-coefficient models have attracted great attention in nonlinear time series analysis recently. In this paper, we consider a semi-parametric functional-coefficient autoregressive model, called the radial basis function network-based state-dependent autoregressive (RBF-AR) model. The stability conditions and existing conditions of limit cycle of the RBF-AR model are discussed. An efficient structured parameter estimation method and the modified multi-fold cross-validation criterion are applied to identify the RBF-AR model. Application of the RBF-AR model to the famous Canadian lynx data is presented. The forecasting capability of the RBF-AR model is compared to those of other competing time series models, which shows that the RBF-AR model is as good as or better than other models for the postsample forecasts.     

Evolutionary learning of dynamic probabilistic models with large time lags

In this paper, we explore the automatic explanation of multivariate time series (MTS) through learning dynamic Bayesian networks (DBNs). We have developed an evolutionary algorithm which exploits certain characteristics of MTS in order to generate good networks as quickly as possible. We compare this algorithm to other standard learning algorithms that have traditionally been used for static Bayesian networks but are adapted for DBNs in this paper. These are extensively tested on both synthetic and real-world MTS for various aspects of efficiency and accuracy. By proposing a simple representation scheme, an efficient learning methodology, and several useful heuristics, we have found that the proposed method is more efficient for learning DBNs from MTS with large time lags, especially in time-demanding situations. © 2001 John Wiley & Sons, Inc.     

Constrained output feedback model predictive control for nonlinear systems

A constrained output feedback model predictive control approach for nonlinear systems is presented in this paper. The state variables are observed using an unscented Kalman filter, which offers some advantages over an extended Kalman filter. A nonlinear dynamic model of the system, considered in this investigation, is developed considering all possible effective elements. The model is then adaptively linearized along the prediction horizon using a state-dependent state space representation. In order to improve the performance of the control system as many linearized models as the number of prediction horizons are obtained at each sample time. The optimum results of the previous sample time are utilized for linearization at the current sample time. Subsequently, a linear quadratic objective function with constraints is formulated using the developed governing equations of the plant. The performance and effectiveness of the proposed control approach is validated both in simulation and through real-time experimentation using a constrained highly nonlinear aerodynamic test rig, a twin rotor MIMO system (TRMS).     

基于周期性建模的时间序列预测方法及电价预测研究

时间序列数据广泛存在于人类的生产生活中, 通常具有复杂的非线性动态和一定的周期性. 与传统的时间序列分析方法相比, 基于深度学习的方法更能捕捉数据的深层特性, 对具有复杂非线性的时间序列有较好的建模效果. 为了在神经网络中显式地建模时间序列数据的周期性和趋势性, 本文在循环神经网络的基础上引入了周期损失和趋势损失, 建立了基于周期性建模和多任务学习的时间序列预测模型. 将模型应用到欧洲能源交易所法国市场的能源市场价格预测中, 结果表明周期损失和趋势损失能够提高神经网络的泛化能力, 并提高预测时间序列趋势的精度.     

基于层次化可塑性回声状态网络的混沌时间序列预测

为了提高回声状态网络对于混沌时间序列特征提取与预测的能力,提出一种层次化可塑性回声状态网络模型.该模型将多个储备池顺序连接,通过逐层特征变换的方式增强对非线性多尺度动态特征的提取能力.同时,引入神经科学中的内在可塑性机制模拟真实生物神经元的放电率分布,以最大化神经元的信息传递为目标对储备池进行预训练.层次化可塑性回声状态网络不仅能够增加模型的容量,降低随机投影所带来的不稳定性,而且也为理解储备池的表示、处理、记忆及储存操作提供一种新的思路.仿真实验结果表明,相比于其他7种改进的回声状态网络模型,所提出的模型在人造数据和真实数据所构成的混沌时间序列预测任务中均能取得最优的预测精度.     

Nonlinear system modeling and robust predictive control based on RBF-ARX model

An integrated modeling and robust model predictive control (MPC) approach is proposed for a class of nonlinear systems with unknown steady state. First, the nonlinear system is identified off-line by RBF-ARX model possessing linear ARX model structure and state-dependent Gaussian RBF neural network type coefficients. On the basis of the RBF-ARX model, a combination of a local linearization model and a polytopic uncertain linear parameter-varying (LPV) model are built to approximate the present and the future system's nonlinear behavior, respectively. Subsequently, based on the approximate models, a min–max robust MPC algorithm with input constraint is designed for the output-tracking control of the nonlinear system with unknown steady state. The closed-loop stability of the MPC strategy is guaranteed by the use of parameter-dependent Lyapunov function and the feasibility of the linear matrix inequalities (LMIs). Simulation study to a NO_x decomposition process illustrates the effectiveness of the modeling and robust MPC approaches proposed in this paper.     

基于时序分解的用电负荷分析与预测

在智能电网普及的大数据背景下,对电力数据进行精准的分析和预测对电网规划和经济部门的管理决策具有重要的指导意义,但大多数模型都只是在单一的时间尺度上进行研究。针对这一问题提出一种基于时序分解的后向传播算法的循环神经网络预测模型。通过对真实的居民用电消费数据以及外部因素数据统计处理,深入地分析了居民用电特点以及行为规律,并根据其数据的特征以及天气、节假日等外部因素对用户用电行为的影响建立预测模型,对用户未来时段的用电量进行预测。此外,考虑到居民用电消费数据的时序特征在不同时间尺度呈现不同的变化规律,通过时序分解建立预测模型来对用户用电行为的周期性和趋势性进行建模,并通过加权融合达到一起训练的效果,具有一定的协同性,提升预测精度。