Consistent dynamic PCA based on errors-in-variables subspace identification

In this paper, we make a comparison between dynamic principal component analysis (PCA) and errors-in-variables (EIV) subspace model identification (SMI) and establish consistency conditions for the two approaches. We first demonstrate the relationship between dynamic PCA and SMI. Then we show that when process variables are corrupted by measurement noise dynamic PCA fails to give a consistent estimate of the process model in general whether or not process noise is present. We then propose an indirect dynamic PCA approach for the consistent estimate of the process model resorting to EIV SMI algorithms. Consistent dynamic PCA models are obtained with and without process disturbances. Additional features of the indirect approach include (i) easy determination of the number of lagged variables in the model; (ii) determination of the number of significant process disturbances; and (iii) consistent estimate of the dynamic PCA models with and without process disturbances. We conduct two simulation examples and an industrial case study to support our theoretical results, where the relationship between dynamic PCA and EIV SMI is numerically verified.     

Identification of errors-in-variables state space models with observation outliers based on minimum covariance determinant

In this paper, a subspace system identification algorithm for the errors-in-variables (EIV) state space models subject to observation noise with outliers has been developed. By using the minimum covariance determinant (MCD) estimator, the outliers have been identified and deleted. Then the classical EIV subspace system identification algorithms have been applied to estimate the parameters of the state space models. In order to solve the MCD problem for the EIV state space models, a random search algorithm has been proposed. A Monte-Carlo simulation results demonstrate the effectiveness of the proposed algorithm.     

Order estimation of multivariable ill-conditioned processes based on PCA method

Ill-conditioned multivariable processes exhibit significantly strong interactions among system variables and large gain directions from the system inputs to the outputs, which makes the identification and control a challenging task. The objective of this paper is to develop an order estimation algorithm for model identification of ill-conditioned processes using subspace methods. In this paper, the order is determined from noise-corrupted samples with high accuracy based on the principal component analysis (PCA) method. To excite each direction in the ill-conditioned process, test signals are designed carefully based on the system characteristics. Using the PCA modeling, the model prediction error is first reconstructed, and the Akaike Information Criterion (AIC) is then used to examine the modeling error bound and thus to determine the process order. A new weighted direction variable is proposed to strengthen the interactions along the small gain directions, thus improving the identifiability and accuracy of the ill-conditioned model. The effectiveness of the proposed method is confirmed by an application study on a high purity distillation column process under noise conditions.     

基于分布式状态空间模型预测控制的多级流程雁阵阵形调整

流程雁阵（process goose queue,PGQ）是一种新颖的流程工业系统分解协调优化方法．针对在过程干扰下多级流程雁阵的阵形调整问题,采用递阶求解的分布式模型预测控制算法,利用输入输出数据的Hankel矩阵,通过子空间辨识方法直接获取流程雁阵的脉冲响应序列,建立了预测控制算法．将此算法应用于一个氧化铝碳酸化分解过程,仿真验证了方法的有效性.     

质子交换膜燃料电池的分数阶非线性状态空间模型研究

针对质子交换膜燃料电池(PEMFC)发电过程中的分数阶和非线性特性,本文提出了一种分数阶子空间辨识方法建立了PEMFC非线性状态空间模型.首先,为了降低建模复杂度,采用典型相关分析法和相关分析法确定了模型输入变量;其次,将分数阶微分理论与Hammerstein模型子空间辨识方法相结合,采用Poisson矩函数对输入输出数据进行预处理,构造了子空间辨识方法的输入输出矩阵,并引入分数阶短时记忆法减少辨识算法计算量;最后,选取多项式作为Hammerstein模型前端静态非线性环节,采用模糊遗传算法优化系统分数阶阶次和系数矩阵.仿真结果验证了算法的有效性,改进的辨识算法可以明显减小计算时间,所得PEMFC辨识模型能够准确地描述PEMFC的动态过程.     

UD分解与偏差补偿结合用于变量带误差模型辨识

本文提出一种基于UD(upper-diagonal)分解与偏差补偿结合的辨识方法,用于变量带误差(errors-in-variables,EIV)模型辨识.考虑单输入单输出(single input and single output,SISO)线性动态系统,当输入和输出含有零均值、方差未知的高斯测量白噪声时,该类系统的模型参数估计是一种典型的EIV模型辨识问题.为了获得这种EIV模型参数的无偏估计,本文先推导出最小二乘模型参数估计偏差量与输入输出噪声方差以及最小二乘损失函数与输入输出噪声方差的关系,然后采用UD分解方法递推获得模型参数估计值,再利用输入输出噪声方差估计值补偿模型参数估计偏差,以此获得模型参数的无偏估计.本文还讨论了算法实现过程中遇到的一些问题及修补方法,并通过仿真例验证了所提辨识方法的有效性.     

Consistency of noise covariance estimation in joint input–output closed-loop subspace identification with application in LQG benchmarking

The LQG trade-off curve has been used as a benchmark for control loop performance assessment. The subspace approach to estimating the LQG benchmark has been proposed in the literature which requires certain intermediate matrices in subspace identification as well as the covariance matrix of the noise. It is shown in this paper that many existing closed-loop identification methods do not give a consistent estimate of the noise covariance matrix. As a result, we propose an alternative subspace formulation for the joint input–output closed-loop identification for which the consistency of the required subspace matrices and noise covariance is guaranteed. Simulation studies and experimental results are provided to demonstrate the utility of the proposed method.     

Orthogonal projection based subspace identification against colored noise

In this paper, a bias-eliminated subspace identification method is proposed for industrial applications subject to colored noise. Based on double orthogonal projections, an identification algorithm is developed to eliminate the influence of colored noise for consistent estimation of the extended observability matrix of the plant state-space model. A shift-invariant approach is then given to retrieve the system matrices from the estimated extended observability matrix. The persistent excitation condition for consistent estimation of the extended observability matrix is analyzed. Moreover, a numerical algorithm is given to compute the estimation error of the estimated extended observability matrix. Two illustrative examples are given to demonstrate the effectiveness and merit of the proposed method.     

A constrained EM algorithm for principal component analysis

We propose a constrained EM algorithm for principal component analysis (PCA) using a coupled probability model derived from single-standard factor analysis models with isotropic noise structure. The single probabilistic PCA, especially for the case where there is no noise, can find only a vector set that is a linear superposition of principal components and requires postprocessing, such as diagonalization of symmetric matrices. By contrast, the proposed algorithm finds the actual principal components, which are sorted in descending order of eigenvalue size and require no additional calculation or postprocessing. The method is easily applied to kernel PCA. It is also shown that the new EM algorithm is derived from a generalized least-squares formulation.     

基于主元分析的子空间辨识算法

子空间辨识算法作为一种优良的多变量系统辨识算法,最近在国内发展很快.但是现在国内介绍的大多数子空间辨识算法在变量有误差(errors-in-variable)时和闭环辨识时辨识结果却是有偏的,这是因为大多数子空间辨识算法都假设输入变量是没有噪声及辨识算法中存在的一个投影过程.文中介绍了一种新的子空间辨识算法,这种算法利用主元分析(PCA)来获取系统矩阵,避免了其他算法中的投影过程,因此该算法在闭环辨识和变量有误差(errors-in-variable)的情况下,辨识结果也是无偏的.最后给出一个仿真例子说明这种辨识算法的辨识效果良好.     

Closed-loop subspace identification using the parity space

It is known that many subspace algorithms give biased estimates for closed-loop data due to the existence of feedback. In this paper we present a new subspace identification method using the parity space employed in fault detection in the past. The basic algorithm, known as subspace identification method via principal component analysis (SIMPCA), gives consistent estimation of the deterministic part and stochastic part of the system under closed loop. Column weighting for SIMPCA is introduced which shows improved efficiency/accuracy. A simulation example is given to illustrate the performance of the proposed algorithm in closed-loop identification and the effect of column weighting.     

An algorithm for identifying symmetric variables in the canonical OR-coincidence algebra system

To simplify the process for identifying 12 types of symmetric variables in the canonical OR-coincidence （COC） algebra system, we propose a new symmetry detection algorithm based on OR-NXOR expansion. By analyzing the relationships between the coefficient matrices of sub-functions and the order coefficient subset matrices based on OR-NXOR expansion around two arbitrary logical variables, the constraint conditions of the order coefficient subset matrices are revealed for 12 types of symmetric variables. Based on the proposed constraints, the algorithm is realized by judging the order characteristic square value matrices. The proposed method avoids the transformation process from OR-NXOR expansion to AND-OR-NOT expansion, or to AND-XOR expansion, and solves the problem of completeness in the dj-map method. The application results show that, compared with traditional methods, the new algorithm is an optimal detection method in terms of applicability of the number of logical variables, detection type, and complexity of the identification process. The algorithm has been implemented in C language and tested on MCNC91 benchmarks. Experimental results show that the proposed algorithm is convenient and efficient.     

有色噪声条件下的子空间辨识改进方法及应用

针对工业生产过程中噪声往往为有色噪声的情况,提出一种改进的子空间辨识方法。传统的子空间辨识方法在系统存在有色噪声时辨识效果不佳,改进方法则采用变换系统模型形式来克服有色噪声对系统的影响,在辨识时直接利用变换系统模型后的数据得到系统较为准确的状态空间模型,实践证明,状态空间模型更适用于工业过程。连续搅拌反应釜（CSTR）系统是一类典型的工业生产系统,将子空间辨识方法应用于CSTR过程的仿真实验,通过比较改进前和改进后的系统预测误差,验证了所提方法的有效性。     

Subspace model identification under load disturbance with unknown transient and periodic dynamics

To overcome the influence from load disturbance with unknown transient and periodic dynamics, as often encountered when performing identification tests in engineering applications, a bias-eliminated subspace model identification method is proposed to realize consistent estimation, which can be used for both open- and closed-loop systems. By decomposing the output response into disturbed and undisturbed components, an oblique projection is subtly introduced to eliminate the disturbance and noise impact so as to obtain unbiased estimation on the deterministic system state matrices, while the disturbance response dynamics could be estimated. In particular, a specific algorithm based on minimizing the output prediction error is given to find out the disturbance period if exists, such that the disturbance effect can be eliminated by the above projection regardless of the disturbance waveform and magnitude. A shift-invariant approach is then given to retrieve the deterministic state matrices. Consistent estimation on the deterministic system matrices is analyzed with a proof. A benmark example from the literature and an industrial injection molding process are used to demonstrate the effectiveness and merit of the proposed method.     

Subspace identification of continuous time models for process fault detection and isolation

This paper proposes a novel subspace approach towards identification of optimal residual models for process fault detection and isolation (PFDI) in a multivariate continuous-time system. We formulate the problem in terms of the state space model of the continuous-time system. The motivation for such a formulation is that the fault gain matrix, which links the process faults to the state variables of the system under consideration, is always available no matter how the faults vary with time. However, in the discrete-time state space model, the fault gain matrix is only available when the faults follow some known function of time within each sampling interval. To isolate faults, the fault gain matrix is essential. We develop subspace algorithms in the continuous-time domain to directly identify the residual models from sampled noisy data without separate identification of the system matrices. Furthermore, the proposed approach can also be extended towards the identification of the system matrices if they are needed. The newly proposed approach is applied to a simulated four-tank system, where a small leak from any tank is successfully detected and isolated. To make a comparison, we also apply the discrete time residual models to the tank system for detection and isolation of leaks. It is demonstrated that the continuous-time PFDI approach is practical and has better performance than the discrete-time PFDI approach.     

基于核主元分析–主元分析的多阶段间歇过程故障监测与诊断

具有过渡特性的多阶段间歇过程故障监测是一个复杂的问题,既需要考虑稳定阶段下的故障监测,也需要考虑不同阶段间的过渡故障监测.为克服传统硬划分方法导致误警和漏报率高的缺陷,同时也为实现更加精确、有效的故障监测与诊断,提出一套完整的基于核主元分析-主元分析(KPCA-PCA)的多阶段间歇过程故障监测与诊断策略.该方法依据数据相似度实现阶段划分,定义模糊隶属度辨识相邻阶段间的过渡,最后对稳定阶段和过渡过程分别建立具有时变协方差的PCA和KPCA故障监测与诊断模型.通过对青霉素发酵过程的仿真平台及工业应用研究表明,该方法具有更可靠的监控性能,能及时、准确的检测出过程中存在的异常情况.     

Multimode process monitoring with PCA mixture model

For multimode processes, Gaussian mixture model (GMM) has been applied to estimate the probability density function of the process data under normal-operational condition in last few years. However, learning GMM with the expectation maximization (EM) algorithm from process data can be difficult or even infeasible for high-dimensional and collinear process variables. To address this issue, a novel multimode process monitoring approach based on PCA mixture model is proposed. First, the PCA technique is directly applied to the covariance matrix of each Gaussian component to reduce the dimension of process variables and to obtain nonsingular covariance matrices. Then the Bayesian Ying-Yang incremental EM algorithm is adopted to automatically optimize the number of mixture components. With the obtained PCA mixture model, a novel process monitoring scheme is derived for fault detection of multimode processes. Three case studies are provided to evaluate the monitoring performance of the proposed method.     

基于子空间辨识的状态空间模型预测控制

针对无法从工业过程中获得准确状态空间模型的问题,提出一种基于子空间辨识的状态空间模型预测控制方法。利用子空间辨识方法得到的状态空间模型作为系统模型,给出约束条件下的预测控制算法。以CD播放器机械臂系统为例,通过状态空间模型预测控制方法实现对系统输出的跟踪控制,仿真结果表明,该方法控制效果良好。     

基于RPCA对高维数据子空间聚类的预测方法

预测子空间聚类PSC算法由于建立在PCA模型下,无法鲁棒地进行主元分析,导致在面对带有强噪声的数据时,聚类性能受到严重影响。为了提高PSC算法对噪声的鲁棒性,利用近年来受到广泛关注的RPCA分解技术得到数据的低秩结构,鲁棒地提取子空间,具体地,通过将RPCA模型融入PSC算法,提出了一种基于RPCA的预测子空间聚类算法。该算法在RPCA模型下检测强影响点,不但可以有效地进行变量选择和模型选择,而且更重要的是改善了PSC算法在噪声环境下的聚类性能。在真实基因表达数据集上的实验结果表明,改进后的算法较之经典的PSC算法无论在无噪声或加噪声环境下都表现出一定聚类优势及良好的鲁棒性。     

Image retrieval based on incremental subspace learning

Many problems in information processing involve some form of dimensionality reduction, such as face recognition, image/text retrieval, data visualization, etc. The typical linear dimensionality reduction algorithms include principal component analysis (PCA), random projection, locality-preserving projection (LPP), etc. These techniques are generally unsupervised which allows them to model data in the absence of labels or categories. In this paper, we propose a semi-supervised subspace learning algorithm for image retrieval. In relevance feedback-driven image retrieval system, the user-provided information can be used to better describe the intrinsic semantic relationships between images. Our algorithm is fundamentally based on LPP which can incorporate user's relevance feedbacks. As the user's feedbacks are accumulated, we can ultimately obtain a semantic subspace in which different semantic classes can be best separated and the retrieval performance can be enhanced. We compared our proposed algorithm to PCA and the standard LPP. Experimental results on a large collection of images have shown the effectiveness and efficiency of our proposed algorithm.