Enhanced Multivariate Process Monitoring for (original) (raw)
Related papers
Improved process monitoring using nonlinear principal component models
International Journal of Intelligent Systems, 2008
This paper presents two new approaches for use in complete process monitoring. The first concerns the identification of nonlinear principal component models. This involves the application of linear principal component analysis (PCA), prior to the identification of a modified autoassociative neural network (AAN) as the required nonlinear PCA (NLPCA) model. The benefits are that (i) the number of the reduced set of linear principal components (PCs) is smaller than the number of recorded process variables, and (ii) the set of PCs is better conditioned as redundant information is removed. The result is a new set of input data for a modified neural representation, referred to as a T2T network. The T2T NLPCA model is then used for complete process monitoring, involving fault detection, identification and isolation.The second approach introduces a new variable reconstruction algorithm, developed from the T2T NLPCA model. Variable reconstruction can enhance the findings of the contribution charts still widely used in industry by reconstructing the outputs from faulty sensors to produce more accurate fault isolation. These ideas are illustrated using recorded industrial data relating to developing cracks in an industrial glass melter process. A comparison of linear and nonlinear models, together with the combined use of contribution charts and variable reconstruction, is presented. © 2008 Wiley Periodicals, Inc.
Nonlinear process monitoring using kernel principal component analysis
Chemical Engineering Science, 2004
In this paper, a new nonlinear process monitoring technique based on kernel principal component analysis (KPCA) is developed. KPCA has emerged in recent years as a promising method for tackling nonlinear systems. KPCA can e ciently compute principal components in high-dimensional feature spaces by means of integral operators and nonlinear kernel functions. The basic idea of KPCA is to ÿrst map the input space into a feature space via nonlinear mapping and then to compute the principal components in that feature space. In comparison to other nonlinear principal component analysis (PCA) techniques, KPCA requires only the solution of an eigenvalue problem and does not entail any nonlinear optimization. In addition, the number of principal components need not be speciÿed prior to modeling. In this paper, a simple approach to calculating the squared prediction error (SPE) in the feature space is also suggested. Based on T 2 and SPE charts in the feature space, KPCA was applied to fault detection in two example systems: a simple multivariate process and the simulation benchmark of the biological wastewater treatment process. The proposed approach e ectively captured the nonlinear relationship in the process variables and showed superior process monitoring performance compared to linear PCA. ?
IFAC Proceedings Volumes, 2009
In this paper Principal Components Analysis (PCA) is used for detecting faults in a simulated wastewater treatment plant (WWTP). Diagnosis tasks are treated using Fisher discriminant analysis (FDA). Both techniques are multivariate statistical techniques used in multivariate statistical process control (MSPC) and fault detection and isolation (FDI) perspectives. PCA reduces the dimensionality of the original historical data by projecting it onto a lower dimensionality space. It obtains the principal causes of variability in a process. If some of these causes change, it can be due to a fault in the process. FDA provides an optimal lower dimensional representation in terms of a discriminant between classes of data, where, in this context of fault diagnosis, each class corresponds to data collected during a specific and known fault. A discriminant function is applied to diagnose faults using data collected from the plant.
Multivariate process monitoring and fault diagnosis by multi-scale PCA
Computers & Chemical Engineering, 2002
Chemical process plant safety, production specifications, environmental regulations, operational constraints, and plant economics are some of the main reasons driving an upward interest in research and development of more robust methods for process monitoring and control. Principal component analysis (PCA) has long been used in fault detection by extracting relevant information from multivariate chemical data. The recent success of wavelets and multi-scale methods in chemical process monitoring and control has catalyzed an interest in the investigation of wavelets based methods for fault detection. In the present work, multi-scale principal component analysis (MSPCA) is used for fault detection and diagnosis. MSPCA simultaneously extracts both, cross correlation across the sensors (PCA approach) and auto-correlation within a sensor (wavelet approach). Using wavelets, the individual sensor signals are decomposed into approximations and details at different scales. Contributions from each scale are collected in separate matrices, and a PCA model is then constructed to extract correlation at each scale. The multi-scale nature of MSPCA formulation makes it suitable to work with process data that are typically non-stationary and represent the cumulative effect of many underlying process phenomena, each operating at a different scale. The proposed MSPCA approach is able to outperform the conventional PCA based approach in detecting and identifying real process faults in an industrial process, and yields minimum false alarms. Additionally, the advantage of MSPCA, over the traditional PCA approach for sensor validation, is also demonstrated on an industrial boiler data set.
2004
Model-based fault detection and isolation (FDI) in plants of complex control systems has been a subject of tremendous research over the last three decades. Most common FDI approaches are based on analytical models of the systems which are often not available in practice for complex multivariate processes. On the other hand, statistical linear correlation models developed using principal component analysis (PCA) can be built from historical operating databases and require no prior knowledge of the plant. Statistical process monitoring (SPM) approaches using these models can easily handle a large number of variables and are very powerful for fault detection. Their main limitations lie in the linear assumption of the process variables and the ability to isolate or diagnose faults. This thesis presents a multivariate statistical process monitoring (MSPM) and fault diagnosis approach based on nonlinear principal component analysis (NLPCA). A technique called NLPCA neural network is applied to model the process of interest. It addresses the linearity limitation of the PCA by assuming that the hidden principal components are nonlinear functions of the observed process variables; therefore, it is more effective in extracting the information from nonlinearly correlated variables than linear methods. A new statistic fault diagnosing scheme is also developed based on analyzing the distribution patterns of the process data in the nonlinear principal component feature space through the use of self-organizing feature mapping (SOFM) and vector quantization algorithms. The proposed procedure is compared with observer-based methods and current statistical methods in performing process monitoring and fault diagnosis of linear and nonlinear processes. Its applications are illustrated on the diesel engine actuator benchmark system as well as the three-tank benchmark system. Dedication I dedicate this thesis to my husband Hui , my lovely son Liam and my parents.
A new multivariate statistical process monitoring method using principal component analysis
Computers & chemical …, 2001
Principal component analysis (PCA) has been used successfully as a multivariate statistical process control (MSPC) tool for detecting faults in processes with highly correlated variables. In the present work, a novel statistical process monitoring method is proposed for further improvement of monitoring performance. It is termed 'moving principal component analysis' (MPCA) because PCA is applied on-line by moving the time-window. In MPCA, changes in the direction of each principal component or changes in the subspace spanned by several principal components are monitored. In other words, changes in the correlation structure of process variables, instead of changes in the scores of predefined principal components, are monitored by using MPCA. The monitoring performance of the proposed method and that of the conventional MSPC method are compared with application to simulated data obtained from a simple 2 × 2 process and the Tennessee Eastman process. The results clearly show that the monitoring performance of MPCA is considerably better than that of the conventional MSPC method and that dynamic monitoring is superior to static monitoring.
Fault diagnosis of Tennessee Eastman process with multi-scale PCA and ANFIS
Chemometrics and Intelligent Laboratory Systems, 2013
Fault diagnosis in industrial processes are challenging tasks that demand effective and timely decision making procedures under the extreme conditions of noisy measurements, highly interrelated data, large number of inputs and complex interaction between the symptoms and faults. The purpose of this study is to develop an online fault diagnosis framework for a dynamical process incorporating multi-scale principal component analysis (MSPCA) for feature extraction and adaptive neuro-fuzzy inference system (ANFIS) for learning the fault-symptom correlation from the process historical data. The features extracted from raw measured data sets using MSPCA are partitioned into score space and residual space which are then fed into multiple ANFIS classifiers in order to diagnose different faults. This data-driven based method extracts fault-symptom correlation from the data eliminating the use of process model. The use of multiple ANFIS classifiers for fault diagnosis with each dedicated to one specific fault, reduces the computational load and provides an expandable framework to incorporate new fault identified in the process. Also, the use of MSPCA enables the detection of small changes occurring in the measured variables and the proficiency of the system is improved by monitoring the subspace which is most sensitive to the faults. The proposed MSPCA-ANFIS based framework is tested on the Tennessee Eastman (TE) process and results for the selected fault cases, particularly those which exhibit highly non-linear characteristics, show improvement over the conventional multivariate PCA as well as the conventional PCA-ANFIS based methods.
Nonlinear system monitoring using multiscaled principal components analysis based on neural network
International Journal of Modelling, Identification and Control
In this paper, we propose a new method based on multiscaled principal component analysis for nonlinear systems analysis. We introduce nonlinear PCA based on neural networks and discrete wavelet transform. The data matrix describing a nonlinear process is decomposed into five wavelet resolution levels. The neural PCA is applied to each coefficient of details and approximations; we select only the scales having a defect to reconstruct the data matrix. Neural PCA is again applied to the new matrix to determine the defective variables, which are detected using the square predictive error (SPE) statistic and identified using the contributions calculation method. This method is applied to a biological process and shows efficient results.
International Journal of Adaptive and Innovative Systems, 2010
In the general frame of process surveillance, principal component analysis (PCA) has been often selected due to its simplicity and ability to capture the linear relations between the stationary process variables. However, the method showed limitations when dealing with industrial data that generally presents non-linear and multiscale features. The approach proposed in this study rests on the modelling using non-linear PCA coupled with artificial neural networks (ANNs) to extract the non-linear inter-correlation between variables and on the wavelet analysis to decompose each sensor signal into a set of coefficients at different scales. The contribution of each variable for each scale is then collected in separated matrices and a non-linear PCA model is constructed for each matrix. The proposed approach is applied to fault detection of pollution parameters affecting the region of Annaba in Algeria. The performance of the approach is then illustrated and compared with those of classic PCA and multiscale PCA (MSPCA).
Monitoring approach using Nonlinear Principal Component Analysis
2011 International Conference on Communications, Computing and Control Applications (CCCA), 2011
This work concerns the principal component analysis applied to the supervision of quality parameters of the flour production line. Our contribution lies in the combined use of the principal component analysis technique and the clustering algorithms in the field of production system diagnosis. This approach allows detecting and locating the system defects, based on the drifts of the product quality parameters. A comparative study between the classification performance by clustering algorithms and the principal component analysis has been proposed. Locating parameters in defect is based on the technique of fault direction in partial least square.