In-Beum Lee

Fault detection of batch processes using multiway kernel principal component analysis

Batch processes are very important in most industries and are used to produce high-quality materials, which causes their monitoring and control to emerge as essential techniques. Several multivariate statistical analyses, including multiway principal component analysis (MPCA), have been developed for the monitoring and fault detection of batch process. In this paper, a new batch monitoring method using multiway kernel principal component analysis (MKPCA) is proposed. Three-way batch data of normal batch process are unfolded batch-wise, and then KPCA is used to capture the nonlinear characteristics within normal batch processes. The proposed monitoring method was applied to fault detection in the simulation benchmark of fed-batch penicillin production. In both off-line analysis and on-line batch monitoring, the proposed approach can effectively capture the nonlinear relationships among process variables. In on-line monitoring, MKPCA can detect significant deviation which may cause a lower quality of final products. MPCA, however, has a limit to detect faults.

Process monitoring using a Gaussian mixture model via principal component analysis and discriminant analysis

Abstract Conventional process monitoring based on principal component analysis (PCA) has been applied to many industrial chemical processes. However, such PCA-based approaches assume that the process is operating in a steady state and consequently that the process data are normally distributed and contain no time correlations. These assumptions significantly limit the applicability of PCA-based approaches to the monitoring of real processes. In this paper, we propose a more exact and realistic process monitoring method that does not require that the process data be normally distributed. Specifically, the concept of conventional PCA is expanded such that a Gaussian mixture model (GMM) is used to approximate the data pattern in the model subspace obtained by PCA. The use of a mixture of local Gaussian models means that the proposed approach can be applied to arbitrary datasets, not just those showing a normal distribution. To use the GMM for monitoring, the overall T2 and Q statistics were used as the monitoring guidelines for fault detection. The proposed approach significantly relaxes the restrictions inherent in conventional PCA-based approaches in regard to the raw data pattern, and can be expanded to dynamic process monitoring without developing a complicated dynamic model. In addition, a GMM via discriminant analysis is proposed to isolate faults. The proposed monitoring method was successfully applied to three case studies: (1) simple two-dimensional toy problems, (2) a simulated 4×4 dynamic process, and (3) a simulated non-isothermal continuous stirred tank reactor (CSTR) process. These application studies demonstrated that, in comparison to conventional PCA-based monitoring, the proposed fault detection and isolation (FDI) scheme is more accurate and efficient.

An enhanced PID control strategy for unstable processes

An enhanced PID control strategy is proposed for unstable processes. Ultimate data sets and the process gain estimated from an improved relay feedback method are used to identify unstable processes as unstable first-order plus time-delay models. We also suggest a simple control structure including an inner feedback loop and the corresponding tuning relations to manipulate unstable processes more efficiently and systematically. The proposed control method not only guarantees the simplicity and the easiness, but also shows much better control performance than previous control methods for unstable processes.

Process monitoring based on probabilistic PCA

Abstract This paper proposes a multivariate process monitoring method based on probabilistic principal component analysis (PPCA). First we will summarize several well-known statistical process monitoring methods, e.g. univariate/multivariate Shewhart charts, and the PCA-based method, i.e. Q and Hotellings T 2 charts. And then the probabilistic method will be proposed and compared to the existing methods. In essence, the univariate Shewhart chart, multivariate Shewhart chart, Q chart, and T 2 chart are unified to the probabilistic method. The PPCA model is calibrated by the expectation and maximization (EM) algorithm similar to PCA by NIPALS algorithm; EM algorithm will be explained briefly in the article. Finally, through an illustrative example, we will show how the probabilistic method works and is applied to the process monitoring.

Nonlinear modeling and adaptive monitoring with fuzzy and multivariate statistical methods in biological wastewater treatment plants.

A new approach to nonlinear modeling and adaptive monitoring using fuzzy principal component regression (FPCR) is proposed and then applied to a real wastewater treatment plant (WWTP) data set. First, principal component analysis (PCA) is used to reduce the dimensionality of data and to remove collinearity. Second, the adaptive credibilistic fuzzy-c-means method is used to appropriately monitor diverse operating conditions based on the PCA score values. Then a new adaptive discrimination monitoring method is proposed to distinguish between a large process change and a simple fault. Third, a FPCR method is proposed, where the Takagi-Sugeno-Kang (TSK) fuzzy model is employed to model the relation between the PCA score values and the target output to avoid the over-fitting problem with original variables. Here, the rule bases, the centers and the widths of TSK fuzzy model are found by heuristic methods. The proposed FPCR method is applied to predict the output variable, the reduction of chemical oxygen demand in the full-scale WWTP. The result shows that it has the ability to model the nonlinear process and multiple operating conditions and is able to identify various operating regions and discriminate between a sustained fault and a simple fault (or abnormalities) occurring within the process data.

New gene selection method for classification of cancer subtypes considering within-class variation

In this work we propose a new method for finding gene subsets of microarray data that effectively discriminates subtypes of disease. We developed a new criterion for measuring the relevance of individual genes by using mean and standard deviation of distances from each sample to the class centroid in order to treat the well‐known problem of gene selection, large within‐class variation. Also this approach has the advantage that it is applicable not only to binary classification but also to multiple classification problems. We demonstrated the performance of the method by applying it to the publicly available microarray datasets, leukemia (two classes) and small round blue cell tumors (four classes). The proposed method provides a very small number of genes compared with the previous methods without loss of discriminating power and thus it can effectively facilitate further biological and clinical researches.

Improved relay auto-tuning with static load disturbance

In the presence of static load disturbance, a relay auto-tuning experiment shows significant errors in estimating the ultimate gain and period. This paper shows how to overcome the problem without any information on the static gain of the process and how to obtain process gain from one relay experiment.

On-line batch process monitoring using a consecutively updated multiway principal component analysis model

Batch processes lie at the heart of many industries; hence the effective monitoring and control of batch processes is crucial to the production of high-quality materials. Multiway principal component analysis (MPCA) has been widely used for batch monitoring and has proved to be an effective method for monitoring many industrial batch processes. However, because MPCA is a fixed-model monitoring technique, it gives false alarms when it is used to monitor real processes whose normal operation involves slow changes. In this paper, we propose a simple on-line batch monitoring method that uses a consecutively updated MPCA model. The key to the proposed approach is that whenever a batch successfully remains within the bounds of normal operation, its batch data are added to the historical database of normal data and a new MPCA model is developed based on the revised database. The proposed method was applied to monitoring fed-batch penicillin production, and the results were compared with those obtained using conventional MPCA. The simulation results clearly show that the ability of the proposed method to adapt to new normal operating conditions eliminates the many false alarms generated by the fixed model and provides a reliable monitoring chart.

Robust PID tuning for Smith predictor in the presence of model uncertainty

Abstract This paper presents robust PID tuning for the Smith predictor in the presence of model uncertainty. The concept of the equivalent gain plus time delay (EGPTD) is introduced to incorporate robust stability in PID tuning of the Smith predictor. In particular, an application is developed for the robust tuning of the first order plus time delay (FOPTD) system and the second order plus time delay (SOPTD) system because the systems have been used extensively to describe chemical processes. The proposed tuning method can cope with simultaneous uncertainties in all parameters of the model in an efficient manner. Another important and attractive feature of the method is that it can utilize many currently available PID tuning rules. Simulation results are provided to demonstrate the availability of the method.

Gene selection and classification from microarray data using kernel machine

The discrimination of cancer patients (including subtypes) based on gene expression data is a critical problem with clinical ramifications. Central to solving this problem is the issue of how to extract the most relevant genes from the several thousand genes on a typical microarray. Here, we propose a methodology that can effectively select an informative subset of genes and classify the subtypes (or patients) of disease using the selected genes. We employ a kernel machine, kernel Fisher discriminant analysis (KFDA), for discrimination and use the derivatives of the kernel function to perform gene selection. Using a modified form of KFDA in the minimum squared error (MSE) sense and the gradients of the kernel functions, we construct an effective gene selection criterion. We assess the performance of the proposed methodology by applying it to three gene expression datasets: leukemia dataset, breast cancer dataset and colon cancer dataset. Using a few informative genes, the proposed method accurately and reliably classified cancer subtypes (or patients). Also, through a comparison study, we verify the reliability of the gene selection and discrimination results.