Dynamic Principal Component Research Articles

Fault Detection and Health Assessment of Equipment Based on Fuzzy DPCA Spatial Eigenvalue Similarity

To improve the fault recognition rate of the dynamic principal component spatial data drive method, a fault diagnosis and equipment health status assessment method based on similarity fuzzy dynamics principal component analysis was proposed. First, the data are fuzzified according to the error function, and an augmented matrix is constructed. The eigenvalues are decomposed to obtain a score matrix and residual matrix of the fuzzy principal component. Further, the similarity between fault data and normal data is calculated. Meanwhile, a health assessment of the equipment is realized. The contribution rate of the observed variables is calculated. Finally, general Tennessee Eastman data and health assessment of a hydraulic press are used to validate the algorithm. The results show that the SFDPCA has a 100 % fault recognition rate for some faults, and the recognition rate for other faults is also higher than that of DPCA-Diss, DPCA-SPE, and PCA-SPE. The SDDPCA accurately identifies abnormal phenomena. It can determine the health level of prefilling and effectively make up for the shortcomings of PCA − T 2 , PCA-SPE, DPCA-Diss, and other methods and also can be applied to data-driven fault diagnosis to improve the fault recognition rate.

Dynamic Non-Gaussian hybrid serial modeling for industrial process monitoring

Process monitoring has been widely used for fault detection and performance supervision in modern industrial processes. Nevertheless, hybrid characteristics including Gaussianity, non-Gaussianity and dynamic usually coexist in process variables, which brings new challenge to obtain satisfactory monitoring performance. Aiming at the hybrid characteristics problem, this paper proposes a dynamic non-Gaussian hybrid serial modeling method for industrial process monitoring. First, a multivariate non-Gaussianity evaluation method is utilized to divide industrial process variables into the Gaussian variable subspace and the non-Gaussian variable subspace. Afterwards considering the hybrid characteristics including Gaussianity, non-Gaussianity and dynamic at information level, a dynamic principal component analysis (DPCA)-dynamic independent component analysis (DICA)-based hybrid serial modeling method is presented for analyzing simultaneously the dynamic Gaussian and non-Gaussian information in each variable subspace. Subsequently, the final monitoring results are obtained using Bayesian inference and the DPCA-DICA-based hybrid serial similarity factor is proposed for fault identification. Unlike the existing methods, the proposed method analyzes simultaneously the Gaussianity, non-Gaussianity and dynamic at different levels of variable and information for improving performance. The case studies including a numerical system, the Tennessee Eastman process and a practical industrial process demonstrate its feasibility and effectiveness.

Dynamic Functional Principal Components for Testing Causality

In this paper, we investigate the causality in the sense of Granger for functional time series. The concept of causality for functional time series is defined, and a statistical procedure of testing the hypothesis of non-causality is proposed. The procedure is based on projections on dynamic functional principal components and the use of a multivariate Granger test. A comparative study with existing procedures shows the good results of our test. An illustration on a real dataset is provided to attest the performance of the proposed procedure.

Bayesian inference based reorganized multiple characteristics subspaces fusion strategy for dynamic process monitoring

The measured data of the large-scale industrial process usually has shown the nonstationary, non-Gaussian, dynamic characteristics, however, most traditional methods did not consider the multiple characteristics coexistence and viewed all the variables as a whole. To make up the deficiencies of the conventional methods, this paper proposes a novel reorganized multiple characteristics subspaces integrated with Bayesian inference (RMS-BI) monitoring strategy for large-scale dynamic process. Firstly, the overall process variables are divided into three subspaces by Jarque–Bera (J–B) test and Augmented Dickey–Fuller (ADF) test, which are the nonstationary subspace, stationary Gaussian subspace, and stationary non-Gaussian subspace. Then, the cointegration analysis (CA), dynamic principal component analysis (DPCA) and dynamic independent component analysis (DICA) models are singled out to monitor the abnormities in the three subspaces, respectively. After that, the monitoring results of the multiple subspaces are integrated by Bayesian inference (BI) to obtain global monitoring statistics. Finally, case studies on the Tennessee Eastman process and the real-world diesel working process are used to demonstrate the availability of the RMS-BI method.

Investigating potential protein markers of cardiovascular disease in children with type 1 diabetes mellitus.

Type 1 diabetes mellitus (T1DM) is a metabolic disease characterized by dysglycaemia. Cardiovascular disease (CVD) is a major complication among T1DM patients and the leading cause of mortality later in life. The study subjects consisted of T1DM children with poor glycemic control (HbA1c>7.5%) and healthy age and gender matched controls. Venous blood samples were collected and tested by utilizing a novel immunoassay panel with 96 protein biomarkers. Data were analyzed using non-linear regression analysis and the expression of biomarkers was compared between T1DM and healthy control groups using an unpaired student's t-test. Dynamic principal component analysis (PCA) was operated based on the differentially expressed proteins. Ten T1DM children and 10 healthy controls were analyzed. Twelve CVD markers show significant differential expression between T1DM patients and healthy controls (p<0.05). Dynamic PCA clustering based on differentially expressed proteins demonstrated an obvious clustering between the two populations. This preliminary study reveals the feasibility of utilizing a novel immunoassay panel to investigate potential biomarkers for predicting incipient CVD in children with T1DM. In future, longitudinal studies are required to track the relationships between measurements of the selected protein markers and the development of CVD in T1DM patients.

Moving dynamic principal component analysis for non-stationary multivariate time series

This paper proposes an extension of principal component analysis to non-stationary multivariate time series data. A criterion for determining the number of final retained components is proposed. An advance correlation matrix is developed to evaluate dynamic relationships among the chosen components. The theoretical properties of the proposed method are given. Many simulation experiments show our approach performs well on both stationary and non-stationary data. Real data examples are also presented as illustrations. We develop four packages using the statistical software R that contain the needed functions to obtain and assess the results of the proposed method.

Dynamic neural orthogonal mapping for fault detection

Dynamic principal component analysis (DPCA) and its nonlinear extension, dynamic kernel principal component analysis (DKPCA), are widely used in the monitoring of dynamic multivariate processes. In traditional DPCA and DKPCA, extended vectors through concatenating current process data point and a certain number of previous process data points are utilized for feature extraction. The dynamic relations among different variables are fixed in the extended vectors, i.e. the adoption of the dynamic information is not adaptively learned from raw process data. Although DKPCA utilizes a kernel function to handle dynamic and (or) nonlinear information, the prefixed kernel function and the associated parameters cannot be most effective for characterizing the dynamic relations among different process variables. To address these problems, this paper proposes a novel nonlinear dynamic method, called dynamic neural orthogonal mapping (DNOM), which consists of data dynamic extension, a nonlinear feedforward neural network, and an orthogonal mapping matrix. Through backpropagation and Eigen decomposition (ED) technique, DNOM can be optimized to extract key low-dimensional features from original high-dimensional data. The advantages of DNOM are demonstrated by both theoretical analysis and extensive experimental results on the Tennessee Eastman (TE) benchmark process. The results on the TE benchmark process show the superiority of DNOM in terms of missed detection rate and false alarm rate. The source codes of DNOM can be found in https://github.com/htz-ecust/DNOM .

Fault Detection and Identification Based on Explicit Polynomial Mapping and Combined Statistic in Nonlinear Dynamic Processes

Single traditional multivariate statistical monitoring methods, such as principal component analysis (PCA) and canonical variate analysis (CVA), are less effective in nonlinear dynamic processes. Monitoring approaches based on radial basis kernel function have been intensively applied. However, an infinite dimension nonlinear mapping is redundant and inefficient. To improve the efficiency of traditional methods and consider the nonlinearity and dynamics simultaneously, this paper proposes canonical variate nonlinear principal component analysis (CV-NPCA) based on explicit polynomial mapping and combined statistic for detecting and identifying faults in nonlinear dynamic processes. There are two main contributions of the proposed method. First, explicit second-order polynomial mapping is introduced to combine CVA with PCA to simultaneously decrease the adverse effects of nonlinearity and dynamics. Second, the $Q_{c}$ statistic combining residual vectors with residual components is proposed, and a two-dimensional (2D) contribution plot and the variable with the largest contribution based on the $Q_{c}$ statistic are given for fault identification in the simulation study. Compared with the results of PCA, CVA, kernel principal component analysis (KPCA), nonlinear dynamic principal component analysis (NDPCA) and kernel entropy component analysis (KECA), the proposed method not only has relatively higher fault detection rates and identification rates but also has lower false alarm rates in the numerical simulation process and the benchmark Tennessee Eastman process.

Abnormal node detection method for time-sharing heating energy consumption in multi-storey buildings based on drosophila algorithm

In order to solve the problem of high energy consumption of time-sharing heating in multi-storey buildings caused by abnormal node, it is necessary to study the detection method of abnormal node of heating energy consumption. Based on fruit flies algorithm of multi-storey building time-sharing heating energy consumption of the abnormal node detection method, using synchronous sequential PLD multi-storey building heating system of heating time-sharing data, put optimisation algorithm is introduced into EEMD algorithm, noise by flies optimisation calculation the corresponding threshold, remove the noise, the IMF component complete data denoising processing. Based on the recursive dynamic principal component analysis method, the detection model of abnormal heating energy consumption nodes is constructed to realise the detection of abnormal heating energy consumption nodes in multi-storey buildings. Experimental results show that the proposed method has good denoising effect, low false alarm rate and high detection rate.

Sparse estimation of dynamic principal components for forecasting high-dimensional time series

We present the sparse estimation of one-sided dynamic principal components (ODPCs) to forecast high-dimensional time series. The forecast can be made directly with the ODPCs or by using them as estimates of the factors in a generalized dynamic factor model. It is shown that a large reduction in the number of parameters estimated for the ODPCs can be achieved without affecting their forecasting performance.

Unsupervised Regression Model of Geodesic Flow Kernel Based on Dynamic Independent Component Analysis and Dynamic Principal Component Analysis

Unsupervised Regression Model of Geodesic Flow Kernel Based on Dynamic Independent Component Analysis and Dynamic Principal Component Analysis

Dynamic Monitoring of Grinding Circuits by Use of Global Recurrence Plots and Convolutional Neural Networks

Reliable control of grinding circuits is critical to more efficient operation of concentrator plants. In many cases, operators still play a key role in the supervisory control of grinding circuits but are not always able to act timely to deal with disturbances, such as changes in the mill feed. Reliable process monitoring can play a major role in assisting operators to take more timely and reliable action. These monitoring systems need to be able to deal with what could be complex nonlinear dynamic behavior of comminution circuits. To this end, a dynamic process monitoring approach is proposed based on the use of convolutional neural networks. To take advantage of the availability of pretrained neural networks, the grinding circuit variables are treated as time series which can be converted into images. Features extracted from these networks are subsequently analyzed in a multivariate process monitoring framework with an underlying principal component model. Two variants of the approach based on convolutional neural networks are compared with dynamic principal component analysis on a simulated and real-world case studies. In the first variant, the pretrained neural network is used as a feature extractor without any further training. In the second variant, features are extracted following further training of the network in a synthetic binary classification problem designed to enhance the extracted features. The second approach yielded nominally better results than what could be obtained with dynamic principal component analysis and the approach using features extracted by transfer learning.

A CPPS based on GBDT for predicting failure events in milling

Cyber-physical production systems (CPPS) are mechatronic systems monitored and controlled by software brains and digital information. Despite its fast development along with the advancement of Industry 4.0 paradigms, an adaptive monitoring system remains challenging when considering integration with traditional manufacturing factories. In this paper, a failure predictive tool is developed and implemented. The predictive mechanism, underpinned by a hybrid model of the dynamic principal component analysis and the gradient boosting decision trees, is capable of anticipating the production stop before one occurs. The proposed methodology is implemented and experimented on a repetitive milling process hosted in a real-world CPPS hub. The online testing results have shown the accuracy of the predicted production failures using the proposed predictive tool is as high as 73% measured by the AUC score.

Time-varying general dynamic factor models and the measurement of financial connectedness

We propose a new time-varying Generalized Dynamic Factor Model for high-dimensional, locally stationary time series. Estimation is based on dynamic principal component analysis jointly with singular VAR estimation, and extends to the locally stationary case the one-sided estimation method proposed by Forni et al. (2017) for stationary data. We prove consistency of our estimators of time-varying impulse response functions as both the sample size T and the dimension n of the time series grow to infinity. This approach is used in an empirical application in order to construct a time-varying measure of financial connectedness for a large panel of adjusted intra-day log ranges of stocks. We show that large increases in long-run connectedness are associated with the main financial turmoils. Moreover, we provide evidence of a significant heterogeneity in the dynamic responses to common shocks in time and over different scales, as well as across industrial sectors.

Analysis of volatile organic compounds (VOCs) fingerprint of raw and honey-fried licorice based on headspace-gas-chromatography ion-mobility spectrometry (HS-GC-IMS)

Licorice is one of the most commonly used traditional Chinese medicine. In clinic, raw licorice and honey-fried licorice are used in medicines, with the main effects in clearing away heat and detoxifying, moistening lungs and removing phlegm. Honey-fried licorice has effects in nourishing the spleen and stomach and replenishing Qi and pulse. Because traditional Chinese medicine exerts the effects through multiple components and multiple targets, the index components used in the quality evaluation of licorice are often difficult to reflect their real quality. In addition, most of studies for the quality standards have shown that honey-fried licorice are the same as licorice, with a lack of quality evaluation standards that can demonstrate their processing characteristics. The quality of medicine is directly related to its clinical efficacy, so it is necessary to establish a more effective quality control method. Licorice has a beany smell, which is one of the main quality identification characteristics. In this study, by taking advantage of the odor characteristics, a headspace-gas chromatography-ion migration mass spectrometry technology was used to establish a quality evaluation method. A total of 76 volatile components were identified. Through the dynamic principal component analysis, 7 kinds of volatile substances in raw licorice and 13 kinds of volatile substances in honey-fried licorice were statistically obtained, and could be taken as index components for the quality evaluation of raw and honey-fried licorice, respectively. This study could help realize the combination and unification of modern detection and traditional quality evaluation methods, and make a more realistic evaluation for the quality of licorice.

The degradation study for QFP interconnection structure based on PCMD health index and Darveaux model

In electronic assemblies, quad flat package (QFP) is a widely used packaging method and its interconnection structure degradation is an important field in electronic reliability research. Although degradation processes and failure modes of QFP are various, presently, there are few researching articles on the degradation regularity and mechanism under the same failure mode. In this paper, QFP degradation experiment has been carried out. Three typical QFP degradation processes and failure modes are obtained by experiment data and crack microscopic analysis. Degradation features and curves are extracted through the health index from fusion of dynamic principal component and Mahalanobis distance(PCMD). The regularity shows that there is a similar inflection point in degradation rate curves, and degradation rates are significantly different after this point. Then, according to Darveaux failure physical model, the strain energy distribution of the crack initiation and propagation area is obtained by finite element simulation. Results show that average strain energy density in crack initiation area is similar but in crack propagation area is not, resulting in crack initiation at similar time but the crack propagation rate is different.

Data-driven multimodal operation monitoring and fault diagnosis of high-speed train bearings

The multiple bearings of high-speed trains operate in a similar condition that correlates the bearings temperatures dynamically. This allows for detecting bearing abnormalities and diagnosing the faults according to unexpected changes in the dynamic relations. Traditional single-modal modeling methods do not consider the different dynamic characteristics while the train is in different operating zones. This leads to difficulty with modeling the relationship using a single model, resulting in false alarms. To solve the above problems, the bearing temperature data collected by the bearing temperature monitoring system, the location information of the historical bearing fault alarms collected by the train control system, and the maintenance data are incorporated to identify the operation modes. In this paper, a novel data-driven multimodal bearing operation monitoring and fault diagnosis method is proposed for high-speed trains. First, to manage the outliers and missing values caused by train network communication disconnection and sensor faults, a preprocessing method, which combines linear-filling and a dynamic principal component pursuit, is proposed to reconstruct the contaminated data of the bearing temperatures. Second, the train location information-based operation mode identification method is proposed according to the fact that the clustering results of multiple bearing temperature correlations match well with the operating zone, extracted from the location information of historical faults. Third, a multimodal dynamic inner canonical correlation analysis (M-DiCCA) modeling method and the corresponding operation monitoring method for the train bearings are proposed. Subsequently, a dynamic time warping (DTW)-based fault diagnosis method is proposed. Finally, the application results by using the bearing temperature data collected from practical train operations demonstrate the effectiveness of the proposed methods.

Regional renewable energy development in China: A multidimensional assessment

Renewable energy (RE) is strategically important to achieve sustainability. Its development in China, however, is clearly imbalanced across regions. The uneven distribution of natural resources, financial resources and other factors across this country has brought serious challenges to the policymakers. It is thus important to give a comprehensive evaluation of China's regional RE development. To do so, this paper introduces a multidimensional approach and establishes a quantitative evaluation framework. Based on the existing literature, five dimensions of factors have been chosen for the framwork: economic foundation, institutions, technological development potential, energy security and environmental protection, and current status of the RE sector. A dynamic principal component analysis technique is applied to data from 29 provinces between 2008 and 2014. The results demonstrate large variations in RE development across provinces in China. More economically developed regions, such as Beijing, Shanghai, and Guangdong, have maintained a higher ranking and have clear advantages in almost all dimensions. Policy implications and recommendations for more balanced development across China are drawn from the empirical results. This approach can also be extended to investigate similar issues in other countries.

Background subtraction in dynamic scenes using the dynamic principal component analysis

This study presents a foreground detection method capable of robustly estimating the background under the presence of dynamic effects. The key contribution of this study is the use of the dynamic principal component analysis to model the serial correlation between successive frames and construct a robust pixel-based background model. The frames are normalised in hue, saturation and value colour space to reduce the effect of illumination changes. To restrict the background model, kernel density estimation is used to identify the distribution of the background time-lagged data matrix and then confidence interval limits are used to determine the corresponding detection thresholds. The foreground is detected using background subtraction. This method is tested on several common sequences such as CDnet 2014, ETSI 2014 and MULTIVISION 2013. The authors also hold comparisons based on quantitative metrics with several state-of-the-art methods. Experimental results show that their method outperforms some state-of-the-art methods and has comparable performance with some depth-based methods.

Dynamic principal component CAW models for high-dimensional realized covariance matrices

We propose a new dynamic principal component CAW model (DPC-CAW) for time-series of high-dimensional realized covariance matrices of asset returns (up to 100 assets). The model performs a spectral decomposition of the scale matrix of a central Wishart distribution and assumes independent dynamics for the principal components' variances and the eigenvector processes. A three-step estimation procedure makes the model applicable to high-dimensional covariance matrices. We analyze the finite sample properties of the estimation approach and provide an empirical application to realized covariance matrices for 100 assets. The DPC-CAW model has particularly good forecasting properties and outperforms its competitors for realized covariance matrices.