No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Quality-Related Dynamic Process Monitoring: Part I
Abstract
PCA, ICA, PLS, etc., are popular MSPC approaches applied in industry process monitoring. In general, both PCA and ICA are employed in process monitoring to identify anomalous.
ResearchGate has not been able to resolve any citations for this publication.
The partial least squares (PLS) method has been widely used in quality-related industrial process monitoring because of its ability to extract quality-related information. Generally, online quality monitoring data cannot be obtained in real time, and in this case, updating the online monitoring model is a serious challenge. In this study, an online monitoring dynamic PLS (OMD-PLS) model that uses the relation between time-delay process data and time-delay quality data is proposed. To accurately monitor the quality-related and process-related fault data, we also propose an online monitoring dynamic concurrent PLS (OMDC-PLS) model based on OMD-PLS, which has the ability to detect slight deviations. Furthermore, an alarm-parameter alarm method based on the OMDC-PLS model is proposed and effectively reduces the false alarm rate (FAR). Finally, numerical simulations and the Tennessee Eastman process (TEP) are used to illustrate the effectiveness of the proposed methods.
Dynamic principal component analysis (DPCA) is generally employed in monitoring dynamic processes and typically incorporates all measured variables. However, for a large-scale process, the inclusion of variables without fault-relevant information may cause redundancy and degrade monitoring performance. In this study, the influence of variable and time-lagged variable selection on the DPCA monitoring performance is analyzed. Then, a fault-relevant performance-driven distributed monitoring scheme is proposed to achieve efficient fault detection and diagnosis. First, performance-driven process decomposition is performed, and the optimal subset of variables and time-lagged variables for each fault are selected through a stochastic optimization algorithm. Second, local DPCA models are established to characterize the process dynamics and generate fault signature evidence. Finally, a Bayesian diagnosis system with the most efficient evidence sources is established to identify the process status. Case studies on a numerical example and the Tennessee Eastman benchmark process demonstrate the efficiency of the proposed monitoring scheme.
In data-based monitoring field, the nonlinear iterative partial least squares procedure has been a useful tool for process data modeling, which is also the foundation of projection to latent structures (PLS) models. To describe the dynamic processes properly, a dynamic PLS algorithm is proposed in this paper for dynamic process modeling, which captures the dynamic correlation between the measurement block and quality data block. For the purpose of process monitoring, a dynamic total PLS (T-PLS) model is presented to decompose the measurement block into four subspaces. The new model is the dynamic extension of the T-PLS model, which is efficient for detecting quality-related abnormal situation. Several examples are given to show the effectiveness of dynamic T-PLS models and the corresponding fault detection methods.
Partial least squares (PLS) regression is widely used to capture the latent relationship between inputs and outputs in static system modeling. Several dynamic PLS algorithms have been proposed to capture the characteristics of dynamic data. However, none of these algorithms provides an explicit expression for the dynamic inner and outer models. In this paper, a dynamic inner PLS algorithm is proposed for dynamic data modeling. The proposed algorithm provides an explicit dynamic inner model that is ensured in deriving the outer model. Several examples are presented to demonstrate the effectiveness of the proposed algorithm.
The development of large interconnected plants has brought the need for the development of active and accurate performance monitoring methods. The commonly used approach to this problem is implementing the multivariate statistics-based process monitoring (MSPM). In MSPM, data modeling methods play the central role in developing normal operation models, based on which the monitoring statistics can be used to track the process operating performance. This paper seeks to perform a comparison study on the commonly used data modeling methods in the MSPM field, including principal component analysis, partial least squares, and canonical correlation analysis, to provide users and practitioners with informative details such that useful guidance can be offered to select the preferable methods. The interconnections between each two of them are first investigated. Then, dynamic extensions based on their static formulations including how they modify and resolve objective functions are revealed. Using the simulated data from the continuous stirred tank reactor benchmark process and real industrial data from the hot strip rolling mill process, parts of theoretical results are validated.
This paper discusses several practical issues of modeling real chemical processes based on process operating data. One of the most important issues of real chemical process modeling is that the input variables are usually correlated, which could cause a serious problem for the standard least squares method. The second issue is that the process data bases are usually large in size, and some of the samples do not offer additional information to other samples. It is most desirable in this case to delete those samples from the data base without loss of information. The third issue is how to select relevant input variables and delete irrelevant variables for a given modeling task. In this paper, we present a systematic approach to dealing with these issues. Firstly, we use cluster analysis to delete samples that carry redundant information to other samples. Secondly, we use the neural net PLS (NNPLS) method (Qin & McAvoy, 1991a) to handle correlation in the data. Thirdly, we propose a relative sensitivity ratio (RSR) table to select relevant input variables and delete irrelevant variables. Finally, the proposed data-based modeling approach is applied to model real chemical processes.
This paper addresses the dynamic non-Gaussian, quality-related fault diagnosis problem for process industries, which is driven by the fact that the quality indices of the industrial products, such as the thickness and flatness in the hot strip mill process (HSMP), are increasingly emphasized. Traditionally, partial least squares (PLS)-based methods are extensively used for quality-related fault diagnosis, however, they are preferred for the static processes. In this paper, a new dynamic PLS model is developed to deal with the quality-related fault diagnosis issue for dynamic processes. In addition, to handle the non-Gaussian property of the dynamic variables, an independent component analysis (ICA) model is successfully combined with the dynamic PLS model. Finally, the proposed method is firstly examined using the Tennessee Eastman process, where it is shown that the new methods perform better that the existing methods. Then they are applied to a real HSMP, where the effectiveness is further convinced from real industrial data.
In this paper, a data-driven dynamic concurrent projection to latent structures (DCPLS) approach is proposed for quality-relevant fault diagnosis of dynamic processes. First, a novel DCPLS algorithm is proposed for dynamic modeling which captures the dynamic correlations between quality variables and process variables. Quality-specific variations, process-specific variations, and quality-process covariations of dynamic processes are monitored respectively. Secondly, a multi-block extension of DCPLS is designed to compute the contributions according to block partition of the lagged variables, in order to help localize faults. Finally, the application results on strip-thickness relevant fault diagnosis for a practical cold rolling continuous annealing process (CAP) demonstrate the effectiveness of the proposed methods.
The issue of quality-related fault detection has attracted much attention in recent years. Partial least squares (PLS) is considered as an efficient tool for predicting and monitoring. However, due to the fact that PLS performs an oblique projection to input space, it is not suitable for quality-related fault detection. On the other hand, PLS is a static method which cannot be used in dynamic systems. In this paper, a dynamic least squares approach is developed by using the structure of auto-regressive moving average exogenous (ARMAX) time-series model. Furthermore, augmented input matrix is decomposed into two orthogonal parts according to their correlations with output, such that quality-related fault detection can be utilized by designing appropriate statistics in two subspaces corresponding to the two parts. The proposed approached is simple and effective for systems with dynamic input and static output, which is a common case for most industrial processes. A numerical example and an industrial process simulator are applied to test the performance of the proposed approach.
This study suggests a hybrid method of signed digraph and partial least squares for the fault diagnosis of chemical processes. Using the local qualitative relationships of each variable in a signed digraph, a process is decomposed into subprocesses to allow for the reliable diagnosis of multiple faults. A partial least-squares model is built for the estimation of each measured variable in each decomposed subprocess. To handle the process dynamics accurately, the input of the partial least-squares model uses both the current values and the past values of the process variables. In each case, the measured and estimated values are compared to diagnose the fault. A case study of a continuously stirred tank reactor illustrates the conclusion that the proposed method greatly improves the accuracy, resolution, and reliability of the diagnosis.
The hybrid fault diagnosis method based on a combination of the signed digraph and partial least-squares (PLS) has the advantage of improving the diagnosis resolution, accuracy, and reliability, compared to those of previous qualitative methods, and of enhancing the ability to diagnose multiple fault [Ind. Eng. Chem. Res. 2003, 42, 6145-6154]. In this study, the method is applied for the multiple fault diagnosis of the Tennessee Eastman challenge process. The target process is decomposed using the local qualitative relationships of each measured variable. Linear and quadratic models based on dynamic PLS are built to estimate each measured variable, which is then compared with the estimated value in order to diagnose the fault. Through case studies, the proposed method demonstrated a good diagnosis capability compared with previous statistical methods.
Partial least squares (PLS) is an efficient tool widely used in multivariate statistical process monitoring. Since standard PLS performs oblique projection to input space $mathbf{X}$, it has limitations in distinguishing quality-related and quality-unrelated faults. Several postprocessing modifications of PLS, such as total projection to latent structures (T-PLS), have been proposed to solve this issue. Further studies have found that these modifications fail to reduce false alarm rates (FARs) of quality-unrelated faults when fault amplitude increases. To cope with this problem, this paper proposes an enhanced quality-related fault detection approach based on orthogonal signal correction (OSC) and modified-PLS (M-PLS). The proposed approach removes variation orthogonal to output space $mathbf{Y}$ from input space $mathbf{X}$ before PLS modeling, and further decomposes $mathbf{X}$ into two orthogonal subspaces in which quality-related and quality-unrelated statistical indicators are designed separately. Compared with T-PLS, the proposed approach has a more robust performance and a lower computational load. Two case studies, including a numerical example and the Tennessee Eastman (TE) process, show the effeteness of the proposed approach.
This paper describes a model of an industrial chemical process for the purpose of developing, studying and evaluating process control technology. This process is well suited for a wide variety of studies including both plant-wide control and multivariable control problems. It consists of a reactor/ separator/recycle arrangement involving two simultaneous gas—liquid exothermic reactions of the following form:A(g) + C(g) + D(g) → G(liq), Product 1,A(g) + C(g) + E(g) → H(liq), Product 2. Two additional byproduct reactions also occur. The process has 12 valves available for manipulation and 41 measurements available for monitoring or control.The process equipment, operating objectives, process control objectives and process disturbances are described. A set of FORTRAN subroutines which simulate the process are available upon request.The chemical process model presented here is a challenging problem for a wide variety of process control technology studies. Even though this process has only a few unit operations, it is much more complex than it appears on first examination. We hope that this problem will be useful in the development of the process control field. We are also interested in hearing about applications of the problem.
Producing value-added products of high-quality is the common objective in industries. This objective is more difficult to achieve in batch processes whose key quality measurements are not available on-line. In order to reduce the variations of the product quality, an on-line batch monitoring scheme is developed based on the multivariate statistical process control. It suggests using the past measured process variables without real-time quality measurement at the end of the batch run. The method, referred to as BDPCA and BDPLS, integrates the time-lagged windows of process dynamic behavior with the principal component analysis and partial least square respectively for on-line batch monitoring. Like traditional MPCA and MPLS approaches, the only information needed to set up the control chart is the historical data collected from the past successful batches. This leads to simple monitoring charts, easy tracking of the progress in each batch run and monitoring the occurrence of the observable upsets. BDPCA and BDPLS models only collect the previous data during the batch run without expensive computations to anticipate the future measurements. Three examples are used to investigate the potential application of the proposed method and make a comparison with some traditional on-line MPCA and MPLS algorithms.
The use of biased least-squares methods for the estimation of pulse-response coefficients in discrete-time process models is proposed. The two specific approaches considered here are partial least squares (PLS) and a method based on the singular value decomposition (SVD). Example applications include the estimation of the open-loop impulse and step responses of a pilot-scale anaerobic wastewater treatment process. The SVD and PLS methods give very similar results for the problems studied. The PLS method typically requires much less computer time, however.
Partial least squares or projection to latent structures (PLS) has been used in multivariate statistical process monitoring similar to principal component analysis. Standard PLS often requires many components or latent variables (LVs), which contain variations orthogonal to Y and useless for predicting Y. Further, the X-residual of PLS usually has quite large variations, thus is not proper to monitor with the Q-statistic. To reduce false alarm and missing alarm rates of faults related to Y, a total projection to latent structures (T-PLS) algorithm is proposed in this article. The new structure divides the X-space into four parts instead of two parts in standard PLS. The properties of T-PLS are studied in detail, including its relationship to the orthogonal PLS. Further study shows the space decomposition on X-space induced by T-PLS. Fault detection policy is developed based on the T-PLS. Case studies on two simulation examples show the effectiveness of the T-PLS based fault detection methods. © 2009 American Institute of Chemical Engineers AIChE J, 2010
This paper provides an overview and analysis of statistical process monitoring methods for fault detection, identification and reconstruction. Several fault detection indices in the literature are analyzed and unified. Fault reconstruction for both sensor and process faults is presented which extends the traditional missing value replacement method. Fault diagnosis methods that have appeared recently are reviewed. The reconstruction-based approach and the contribution-based approach are analyzed and compared with simulation and industrial examples. The complementary nature of the reconstruction- and contribution-based approaches is highlighted. An industrial example of polyester film process monitoring is given to demonstrate the power of the contribution- and reconstruction-based approaches in a hierarchical monitoring framework. Finally we demonstrate that the reconstruction-based framework provides a convenient way for fault analysis, including fault detectability, reconstructability and identifiability conditions, resolving many theoretical issues in process monitoring. Additional topics are summarized at the end of the paper for future investigation. Copyright © 2003 John Wiley & Sons, Ltd.
The issue of modeling and control of multivariable chemical process systems using the dynamic version of a popular multivariate statistical technique, namely, projection to latent structures (partial least squares or PLS) is addressed. Discrete input-output data are utilized to construct a projection-based dynamic model that captures the dominant features of the process under study. The structure of the resulting model enables the synthesis of a multiloop control system. In addition, the design of feedforward control for multivariable systems using the dynamic PLS framework is also presented. Three case studies are used to illustrate the modeling and control of multivariable linear and nonlinear systems using the suggested approach.
Industrial processes usually involve a large number of variables,
many of which vary in a correlated manner. To identify a process model
which has correlated variables, an ordinary least squares approach
demonstrates ill-conditioned problem and the resulting model is
sensitive to changes in sampled data. In this paper, a recursive partial
least squares (PLS) regression is used for online system identification
and circumventing the ill-conditioned problem. The partial least squares
method is used to remove the correlation by projecting the original
variable space to an orthogonal latent space. Application of the
proposed algorithm to a chemical processing modeling problem is
discussed
In this paper we extend previous work by ourselves and other researchers in the use of principal component analysis (PCA) for statistical process control in chemical processes. PCA has been used by several authors to develop techniques to monitor chemical processes and detect the presence of disturbances [1–5]. In past work, we have developed methods which not only detect disturbances, but isolate the sources of the disturbances [4]. The approach was based on static PCA models, T2 and Q charts [6], and a model bank of possible disturbances. In this paper we use a well-known ‘time lag shift’ method to include dynamic behavior in the PCA model. The proposed dynamic PCA model development procedure is desirable due to its simplicity of construction, and is not meant to replace the many well-known and more elegant procedures used in model identification. While dynamic linear model identification, and time lag shift are well known methods in model building, this is the first application we are aware of in the area of statistical process monitoring. Extensive testing on the Tennessee Eastman process simulation [7] demonstrates the effectiveness of the proposed methodology.
A modification of the Partial Least-Squares (PLS) modelling procedure is presented which permits incorporation of a dynamic transformation into the standard algebraic form of this model. This makes it possible to use the resulting dynamic PLS model for control system design by employing precompensators and postcompensators constructed from the input and output loading matrices and basing the controller for the compensated plant on the dynamic inner relation of the modified PLS model.