Subspace identification of closed loop systems by the orthogonal decomposition method

In this paper, we consider a problem of identifying the deterministic part of a closed loop system by applying the stochastic realization technique of (Signal Process. 52 (2) (1996) 145) in the framework of the joint input-output approach. Using a preliminary orthogonal decomposition, the problem is reduced to that of identifying the plant and controller based on the deterministic component of the joint input-output process. We discuss the role of input signals in closed loop identification and the realization method based on a finite data, and then sketch a subspace method for identifying state space models of the plant and controller. Since the obtained models are of higher order, a model reduction procedure should be applied for deriving lower order models. Some numerical results are included to show the applicability of the present technique.     

脉冲响应模型的闭环子空间辨识方法

针对实际情况中难以采集控制系统开环数据的问题，对脉冲响应模型的闭环辨识问题进行了研究，提出了一种基于正交分解理论的脉冲响应模型闭环子空间辨识方法。通过使用正交分解得到联合输入输出信号的确定部分（Deterministic Components），把闭环问题转化成开环问题。利用Toeplitz矩阵下三角结构形式，对由脉冲响应模型系数组成的子空间矩阵分块分解，通过求取子空间矩阵元素的平均值来获取脉冲响应模型参数的估计。通过采用PID控制器的单输入单输出（SISO）数值仿真、多输入多输出（MIMO）数值仿真和Wood-Berry蒸馏过程仿真实验3个仿真实例，对比研究了所提方法与PBSID_OPT、CVA三种方法。仿真结果表明了所提出的方法具有良好的辨识性能。对于实际工业过程的建模问题，该研究所提的闭环子空间辨识方法具有实际的参考价值和一定的指导意义。     

基于小波包分析的虹膜特征提取方法

提出了一种新的虹膜特征提取方法。针对虹膜纹理图像的特征表示主要集中在中高频能量部分,对图像进行二级小波包分解,并提取出第一级和第二级的中高频部分的能量分量作为特征向量,并采用欧几里德距离作为模式识别方法。克服了用小波分解只提取虹膜的低频部分,从而不能充分地反映纹理的特征的缺陷。试验结果和数据表明了此算法的合理性和有效性。     

基于正交分解的递推子空间闭环辨识方法

为实现闭环系统在线辨识,提出递推正交分解闭环子空间辨识方法(RORT)。首先,根据闭环系统状态空间模型和数据间投影关系,构建确定-随机模型,并利用GIVENS变换实现投影向量的递推QR分解;然后,引入带遗忘因子的辨识算法,构建广义能观测矩阵的递推更新形式,以减少子空间辨识算法中QR分解和SVD分解的计算量;最后,针对某型号陀螺仪闭环系统进行实验。实验结果表明, RORT法的辨识拟合度高于91%,能够对陀螺仪闭环系统模型参数进行在线监测。     

Decomposing time series into deterministic and stochastic influences: A survey

Temporal data produced by industrial, human, and natural phenomena typically contain deterministic and stochastic influences, being the first ideally modelled using Dynamical Systems while the second is appropriately addressed using Statistical tools. Although such influences have been widely studied as individual components, specific tools are required to support their decomposition for a proper modeling and analysis. This article addresses a comprehensive survey of the main time-series decomposition strategies and their relative performances in different application domains. The following strategies are discussed: i) Fourier Transform, ii) Wavelet transforms, iii) Moving Average, iv) Singular Spectrum Analysis, v) Lazy, vi) GHKSS, and vii) other approaches based on the Empirical Mode Decomposition method. In order to assess these strategies, we employ diverse and complementary performance measures: i) Mean Absolute Error, Mean Squared and Root Mean Squared Errors; ii) Minkowski Distances; iii) Complexity-Invariant Distance; iv) Pearson correlation; v) Mean Distance from the Diagonal Line; and vi) Mean Distance from Attractors. Each decomposition strategy is better devoted to particular scenarios, however, without any previous knowledge on data, GHKSS confirmed to work as a fair and general baseline besides its time complexity.     

分布参数系统的时空ARX建模及预测控制

本文针对一类可由抛物型偏微分方程描述的分布参数系统,研究了一种基于输入输出数据的建模与控制方法.首先利用Karhunen-Loève（K-L）分解提取系统的一组主导空间基函数,并以此对系统输出进行时空分解,随后由时空分解得到的时间系数部分以及系统激励构成输入输出信息,利用最小二乘法辨识出时域ARX模型,最后针对该模型设计了广义预测控制器.仿真结果表明,上述控制方法能够对分布参数系统取得良好的控制效果.     

Partial ensemble approach to resolve the mode mixing of extreme-point weighted mode decomposition

Empirical mode decomposition (EMD) is an effective tool for breaking down components (modes) of a nonlinear and non-stationary signal. Recently, a newly adaptive signal decomposition method, namely extreme-point weighted mode decomposition (EWMD), was put forward to improve the performance of EMD, in particular, to resolve the over- or undershooting issue associated with the large amplitude variations. However, similar to EMD, EWMD also suffers the mode mixing problem caused by intermittence or noisy signals. In this paper, inspired by complementary ensemble EMD (CEEMD), a noise-assisted data analysis method called partial ensemble extreme-point weighted mode decomposition (PEEWMD) is proposed to eliminate the mode mixing problem and enhance the performance of EWMD. In the proposed PEEWMD method, firstly white noises in pairs are added to the targeted signal and then the noisy signals are decomposed using the EWMD method to obtain the intrinsic mode functions (IMFs) in the first several stages. Secondly, permutation entropy is employed to detect the components that cause mode mixing. The residual signal is obtained after the identified components are separated from the original signal. Lastly, the residual signal is fully decomposed by using the EWMD method. The proposed PEEWMD method is compared with original EWMD, ensemble EWMD (EEWMD) and CEEMD using simulated signals. The results demonstrate that PEEWMD can effectively restrain the mode mixing issue and generates IMFs with much better performance. Based on that the PEEWMD and envelope power spectrum based fault diagnosis method was proposed and applied to the rubbing fault identification of rotor system and the fault diagnosis of rolling bearing with inner race. The result indicates that the proposed method of fault diagnosis gets much better effect than EMD and EWMD.     

融合序列分解与Prophet模型的时序预测

传统时序预测方法其预测过程无法在相同数据集上推出共享模式, 而机器学习方法无法较好地处理非线性和大规模数据集, 并且需要手动设计特征工程. 深度学习方法弥补了传统预测方法需要高计算高人力的弊端, 用自动学习特征工程代替了手动设计特征工程. 但仅使用深度学习的预测方法所作结构假设较少, 通常需要较高的计算资源以及大量的数据来学习得到准确的模型. 针对上述问题, 本文提出通过采用融合t检验的EMD经验模态将序列分为高频分量和低频分量, 对高频分量使用传统STL序列分解方法进一步对数据做处理, 对高频、低频分量分别进行Prophet预测. 实验结果表明, 相较于传统的LSTM以及Prophet预测模型, 经过STL序列分解后的周期数据能够提升模型的整体预测精确度而融合EMD经验模态的Prophet模型则大大提升了训练效率.     

A decomposition method with minimum communication amount for parallelization of multi-dimensional FFTs

The fast Fourier transform (FFT) is undoubtedly an essential primitive that has been applied in various fields of science and engineering. In this paper, we present a decomposition method for the parallelization of multi-dimensional FFTs with the smallest communication amounts for all ranges of the number of processes compared to previously proposed methods. This is achieved by two distinguishing features: adaptive decomposition and transpose order awareness. In the proposed method, the FFT data is decomposed based on a row-wise basis that maps the multi-dimensional data into one-dimensional data, and translates the corresponding coordinates from multi-dimensions into one dimension so that the one-dimensional data can be divided and allocated equally to the processes using a block distribution. As a result and different from previous works that have the dimensions of decomposition pre-defined, our method can adaptively decompose the FFT data on the lowest possible dimensions depending on the number of processes. In addition, this row-wise decomposition provides plenty of alternatives in data transpose, and different transpose order results in different amounts of communication. We identify the best transpose orders with the smallest communication amounts for the 3-D, 4-D, and 5-D FFTs by analyzing all possible cases. We also develop a general parallel software package for the most popular 3-D FFT based on our method using the 2-D domain decomposition. Numerical results show good performance and scaling properties of our implementation in comparison with other parallel packages. Given both communication efficiency and scalability, our method is promising in the development of highly efficient parallel packages for the FFT.     

A finite element formulation for dynamic parameter identification of robot manipulators

In this paper a finite element based approach is described for the automatic generation of models suitable for dynamic parameter identification. The method involves a nonlinear finite element formulation in which both links and joints are considered as specific finite elements [6, 7]. Since the identification procedure considers rigid-link robot models, the inertial properties of the link elements are described using a lumped mass formulation. The parameters to be identified are masses, first-order moments and inertial tensor components of the links. The equations of motion are written in a form which is linear in the dynamic parameters. This formulation is obtained by employing Jourdain’s principle of virtual power. The parameters are estimated using a linear least squares technique. Singular value decomposition of the regression matrix is used to find the minimum parameter set. Simulation results obtained from the 6 DOF PUMA 560 robot based on the estimated parameters show that the method yields accurate responses.     

基于分解集成的航空货运需求区间预测研究

航空货运是国家重要的战略资源,在国内及国际间的贸易中扮演着不可或缺的角色。对航空货运需求进行的科学预测是航空公司制定基础设施规划和总体投资决策的重要依据。针对航空货运量数据的不确定性,从实际需求出发,引入Bootstrap方法进行不确定性估计,提出一种基于分解集成的区间预测方法。具体来说,首先用局部加权回归的时间序列分解（STL）方法将货运需求数据进行分解。其次,由支持向量回归（SVR）和季节自回归综合移动平均（SARIMA）分别预测分解所得的趋势分量与季节分量。再次,创新性地将白噪声分量进行提取并用Bootstrap方法做重采样处理。最后,将预测结果与处理后的白噪声进行集成重构,利用分位数构造区间进行不确定性量化。对中国两大枢纽机场货运数据的实验结果表明,构建的区间能够有效地结合预测结果量化不确定性,为区间预测提供了一种新的研究思路。     

基于二维经验模式分解的医学图像融合方法

基于多尺度多分辨率的图像融合是医学图像融合的重要方法,二维经验模式分解（BEMD）方法是一种新的多尺度多分辨率图像分解方法. 本文提出了一种基于BEMD的医学图像融合方法. 首先将待融合的两幅图像进行BEMD分解,获得多个BIMF分量和一个剩余分量;然后针对BIMF分量和剩余分量采用不同的融合规则进行图像融合;最后对融合后的各分量进行BEMD逆变换,得到最终的融合结果. 实验结果表明,本文方法可得到较好的融合效果,融合图像清晰,含有的更多信息.     

基于小波分解的网络流量时间序列建模与预测

提出一种基于小波分解的网络流量时间序列的分析和预测方法。将非平稳的网络流量时间序列通过小波分解成为多个平稳分量,采用自回归滑动平均方法分别对各平稳分量进行建模,将所有分量的模型进行组合,得到原始非平稳网络流量时间序列的预测模型。在仿真实验中,利用网络流量文库的时间序列数据建立了预测模型,并对其进行独立测试检验。仿真结果表明,本预测方法提高了网络流量时间序列的预测准确率,是一种有效、稳健的网络流量预测方法。     

The complex local mean decomposition

The local mean decomposition (LMD) has been recently developed for the analysis of time series which have nonlinearity and nonstationarity. The smoothed local mean of the LMD surpasses the cubic spline method used by the empirical mode decomposition (EMD) to extract amplitude and frequency modulated components. To process complex-valued data, we propose complex LMD, a natural and generic extension to the complex domain of the original LMD algorithm. It is shown that complex LMD extracts the frequency modulated rotation and envelope components. Simulations on both artificial and real-world complex-valued signals support the analysis.     

经验模式分解与时间序列分析在网络流量预测中的应用

提出一种经验模式分解和时间序列分析的网络流量预测方法. 首先,对网络流量时间序列进行经验模式分解,产生高低频分量和余量;然后,对各分量进行时间序列分析,确保高频分量采用改进和声搜索算法优化的最小二乘支持向量机模型、低频分量和余量采用差分自回归滑动平均模型进行建模和预测;最后,将预测结果通过RBF神经网络进行非线性叠加,得到最终的预测值.仿真实验表明,所提出方法具有更好的预测效果和更高的预测精度.     

基于ＥＥＭＤ域统计模型的话音激活检测算法

提出了一种基于EEMD域统计模型的话音激活检测算法。算法首先利用总体平均经验模态分解(Ensemble empirical mode decomposition,EEMD)对带噪语音进行分解,得到信号的本征模式函数(Intrinsicmode function,IMF)分量,选择与原信号的相关性最高的两个分量相加组成主分量;然后对主分量进行频域分解,引入统计模型,求出EEMD域特征参数;最后利用噪声与语音的EEMD域特征参数的不同来进行语音激活检测。实验结果表明,在不同信噪比情况下,本文算法性能优于目前常用的VAD算法,特别在噪声强度大时体现出明显的优势。     

基于CEEMDAN的MEMS加速度计输出信号降噪方法

针对微机电系统(MEMS)加速度计输出信号存在误差,导致高压输电杆塔倾斜监测系统的输出倾角数据精确度不高的问题,提出了一种基于自适应噪声完备集合经验模态分解(CEEMDAN)联合奇异值分解(SVD)对杆塔的加速度计输出信号降噪方法。利用CEEMDAN对原始加速度计输出信号进行分解,得到一系列模态分量,分别计算其排列熵(PE),筛选出特征分量和含噪特征分量,然后再将需进一步降噪的特征分量通过SVD进行二次滤波,最后将降噪后的特征分量与未处理的特征分量进行叠加即得到降噪后的加速度计输出信号。仿真实验结果表明,所提出的方法可以有效地抑制噪声干扰,通过与扩展卡尔曼滤波和CEEMDAN-PE对比说明该方法滤波效果更好,有效提高了加速度信号分析精度和杆塔倾斜角测量精度。     

An efficient approach to approximate solutions of eighth-order boundary-value problems

In this paper, the approximate solutions to the eighth-order boundary-value differential equations are solved by using the Adomian decomposition method (ADM). The numerical solutions of the problem are calculated in the form of a series with easily computable components. The numerical illustrations show that this technique is more reliable, efficient and accurate than the traditional schemes.     

基于矩阵分解的海洋SAR图像舰船检测

针对海洋原始图像与低秩和稀疏矩阵分解模型数据结构不一致的问题,本文提出一种新的基于矩阵分解的海洋SAR图像舰船检测方法。首先该方法需对结构化相似的海洋SAR图像进行重组;然后根据重组矩阵特性适应性设计一个分解精度更高、分解速度更快的新矩阵分解模型,并利用增广拉格朗日乘子法求解模型,在不依赖任何杂波模型和检测统计量的前提下,实现代表舰船目标的稀疏成分的提取;最后利用形态学处理进行优化,实现海洋SAR图像舰船目标的检测。基于高分三号SAR卫星数据的实验结果表明,相比已有的基于鲁棒主成分分析的舰船检测方法,本文方法在处理复杂海况时,能更快速度地以较好的形状从海杂波中准确提取舰船目标,具有更好的鲁棒性。     

基于张量分解的分布式主题分类模型

针对大规模数据分类时计算时间长以及分类精度下降等问题,提出使用张量分解求解LDA主题模型参数,实现对海量网络数据的采集、分类、挖掘.该方法使用矩量法将LDA模型求解转化为低维的张量分解问题,通过分解和反射进行参数的传递,运用大数据平台Spark的进行分布式计算.实验结果表明,改进的模型参数计算方法在时间效率和困惑度方面都得到了提升,并且分类信息更加直观,更加适用于大规模网络数据分类工作.