Soluble microbial products (SMP) in the sludge water phase are regarded as the main foulant in MB... more Soluble microbial products (SMP) in the sludge water phase are regarded as the main foulant in MBRs. This study further developed an existing hydrodynamic model by incorporating energy consumption. The focus was on the cost-effectiveness of crossflow (CF) velocity in the control of submicron particle deposition. A sensitivity analysis showed that CF had the greatest impact on both particle backtransport and energy consumption. The other operational variables, i.e., dry solid content (DS), membrane tube dimension (D and L) and temperature (T) were generally less influential with respect to particle backtransport and energy consumption. Submicron particles were likely to deposit in side-stream MBRs, and the lowest backtransport velocity was found for particle radii around 0.1 m and CF below 0.5 m/s. A particle size distribution (PSD) profile of MBR sludge showed a main peak at 40 m and a second peak at 0.1-1 m. The abundance of submicron particles at 2000 kDa was confirmed by a Liquid chromatography-Organic Carbon Detection (LC-OCD) analysis. The colloids responsible for the second peak in the PSD received high weighting factors (high filter cake formation potential) in the model optimization. In a lab-scale MBR, this critical crossflow velocity was between 0.75 and 1 m/s at 40 L/(m 2 h).
Application of conventional statistical monitoring methods to periodic processes can result in fr... more Application of conventional statistical monitoring methods to periodic processes can result in frequent false alarms and/or missed faults due to their common non-stationary behavior seen over a period. To address this, we propose to identify and use a stochastic statespace model that describes statistical behavior of the changes occurring from period to period. This model, when retooled as a periodically time-varying model, can be used for on-line monitoring and estimation with the aid of a Kalman filter. The same model can also be used for inferential estimation of the variables that ere difficult or slow to measure on-line. The proposed approach is applied to a simulation benchmark of waste-water treatment process, which exhibit strong diurnal changes in the feed stream, and compared against the Principal Component Analysis (PCA) and and Partial Least Squares (PLS) methods.
... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear m... more ... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear mapping )(⋅ Φ from an ... feature space. As a result, KPCA performs a nonlinear PCA in the input space (Romdhani et al., 1999). If ...
Batch processes lie at the heart of many industries; hence the effective monitoring and control o... more 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. #
DNA microarray technologies, which monitor simultaneously, the expression pattern of thousands of... more DNA microarray technologies, which monitor simultaneously, the expression pattern of thousands of individual genes in different biological systems have resulted in a tremendous increase of the amount of available gene expression data and have provided new insights into gene expression during development, within disease processes, and across species. However, microarray gene expression data are characterized by very high dimensionality (genes), relatively small numbers of samples (observations), irrelevant features, as well as collinear and multivariate characteristics. These features complicate the interpretation and analysis of microarray data, and the complexity of such data means that its analysis entails a high computational cost. This situation motivated the researchers to develop a new method for analyzing microarray data. In this paper, we propose a simple gene selection and multivariate fuzzy statistical analysis methods. The proposed method was applied to microarray data from leukemia patients; specifically, it was used to interpret the gene expression pattern and analyze the leukemia subtype whose expression profiles correlated with four cases of acute leukemia gene expression.
This article considers new and existing technologies for water reuse networks for water and waste... more This article considers new and existing technologies for water reuse networks for water and wastewater minimization. For the systematic design of water reuse networks, the theory of the water pinch methodology and the mathematical optimization are described, which are proved to be effective in identifying water reuse opportunities. As alternative solutions, evolutionary solutions and stochastic design approaches to water system design are also illustrated. And the project work flow and an example in a real plant are examined. Finally, as development is in the forefront in process industries, this paper will also explore some research challenges encountered in this field such as simultaneous water and energy minimization, energy-pinch design, and eco-industrial parks (EIP).
We propose a new algorithm for adaptive control and self tuning control, referred to as the gener... more We propose a new algorithm for adaptive control and self tuning control, referred to as the generalized damped least squares (GDLS) algorithm. This algorithm is constructed by adding a multi-step penalty for parameter variations to the objective function of the normal least squares algorithm to prevent the singularity problem that leads to estimation windup. We show that the proposed method has properties almost equivalent to those of the normal least squares method, which guarantees that the proposed algorithm is suitable for poorly excited situations. Simulation results show that the proposed method gives better estimation performance than previous methods in spite of its simplicity. The proposed method also shows good parameter tracking performance and no estimation windup. #
The goal of this work is the development of a suitable monitoring module, which is to be the firs... more The goal of this work is the development of a suitable monitoring module, which is to be the first module of an integrated fault detection and control system for the SHARON process. To model the process properly, different PCA models are tested. As a first step, PCA is used in an iterative manner to exclude data not considered to represent normal operational conditions and process behaviour from the original data set. To improve the performance of the identified model, it is decided to account for dynamics in the SHARON process by means of auto-regressive exogenous (ARX) structuring of data before the identification. A fruitful replacement of missing values for this purpose is done by means of a static PCA model. It is shown that the different criteria used in model selection lead to the same DPCA model. In this paper all steps of the monitoring module design are explained and the performance of different models is analyzed.
Most multivariate statistical monitoring methods based on principal component analysis (PCA) assu... more Most multivariate statistical monitoring methods based on principal component analysis (PCA) assume implicitly that the observations at one time are statistically independent of observations at past time and the latent variables follow a Gaussian distribution. However, in real chemical and biological processes, these assumptions are invalid because of their dynamic and nonlinear characteristics. Therefore, monitoring charts based on conventional PCA tend to show many false alarms and bad detectability. In this paper, a new statistical process monitoring method using dynamic independent component analysis (DICA) is proposed to overcome these disadvantages. ICA is a recently developed technique for revealing hidden factors that underlies sets of measurements followed on a non-Gaussian distribution. Its goal is to decompose a set of multivariate data into a base of statistically independent components without a loss of information. The proposed DICA monitoring method is applying ICA to the augmenting matrix with time-lagged variables. DICA can show more powerful monitoring performance in the case of a dynamic process since it can extract source signals which are independent of the auto-and cross-correlation of variables. It is applied to fault detection in both a simple multivariate dynamic process and the Tennessee Eastman process. The simulation results clearly show that the method e ectively detects faults in a multivariate dynamic process. ?
Batch processes are very important in most industries and are used to produce high-quality materi... more 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.
In this paper, a new nonlinear process monitoring technique based on kernel principal component a... more 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. ?
In this paper we propose a new statistical method for process monitoring that uses independent co... more In this paper we propose a new statistical method for process monitoring that uses independent component analysis (ICA). ICA is a recently developed method in which the goal is to decompose observed data into linear combinations of statistically independent components . Such a representation has been shown to capture the essential structure of the data in many applications, including signal separation and feature extraction. The basic idea of our approach is to use ICA to extract the essential independent components that drive a process and to combine them with process monitoring techniques. I 2 , I 2 e and SPE charts are proposed as on-line monitoring charts and contribution plots of these statistical quantities are also considered for fault identification. The proposed monitoring method was applied to fault detection and identification in both a simple multivariate process and the simulation benchmark of the biological wastewater treatment process, which is characterized by a variety of fault sources with non-Gaussian characteristics. The simulation results clearly show the power and advantages of ICA monitoring in comparison to PCA monitoring.
This study shows that an MBR pilot plant with UCT configuration is able to obtain high nutrient r... more This study shows that an MBR pilot plant with UCT configuration is able to obtain high nutrient removal efficiency already during start-up. The biological nutrient removal (BNR) efficiencies significantly increased towards the end of the experimental run, achieving a COD removal efficiency exceeding 94% and N removal efficiency in the range of 89 to 93%. P removal efficiencies in the range of 80 to 92% have been obtained. During the experimental period (4 months) the evolution of the activity of polyphosphate-accumulating organisms, obtained from P release and P uptake rates, showed a small increase in the activity of polyphosphateaccumulating organisms (PAOs) and denitrifying polyphosphate-accumulating organisms (DPAOs). The specific phosphate accumulation at the end of the experimental run amounted to 8.0 mg P g − 1 VSS h − 1 and 3.29 mg P g − 1 VSS h − 1 , for the PAOs and DPAOs respectively. Moreover, the DPAOs activity increased faster than PAOs activity, i.e. from 0.36 to 0.41 of phosphate uptake rate (PUR) ratio.
Application of conventional statistical monitoring methods to periodic processes can result in fr... more Application of conventional statistical monitoring methods to periodic processes can result in frequent false alarms and/or missed faults due to their common non-stationary behavior seen over a period. To address this, we propose to identify and use a stochastic statespace model that describes statistical behavior of the changes occurring from period to period. This model, when retooled as a periodically time-varying model, can be used for on-line monitoring and estimation with the aid of a Kalman filter. The same model can also be used for inferential estimation of the variables that ere difficult or slow to measure on-line. The proposed approach is applied to a simulation benchmark of waste-water treatment process, which exhibit strong diurnal changes in the feed stream, and compared against the Principal Component Analysis (PCA) and and Partial Least Squares (PLS) methods.
... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear m... more ... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear mapping )(⋅ Φ from an ... feature space. As a result, KPCA performs a nonlinear PCA in the input space (Romdhani et al., 1999). If ...
Batch processes lie at the heart of many industries; hence the effective monitoring and control o... more 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. #
Soluble microbial products (SMP) in the sludge water phase are regarded as the main foulant in MB... more Soluble microbial products (SMP) in the sludge water phase are regarded as the main foulant in MBRs. This study further developed an existing hydrodynamic model by incorporating energy consumption. The focus was on the cost-effectiveness of crossflow (CF) velocity in the control of submicron particle deposition. A sensitivity analysis showed that CF had the greatest impact on both particle backtransport and energy consumption. The other operational variables, i.e., dry solid content (DS), membrane tube dimension (D and L) and temperature (T) were generally less influential with respect to particle backtransport and energy consumption. Submicron particles were likely to deposit in side-stream MBRs, and the lowest backtransport velocity was found for particle radii around 0.1 m and CF below 0.5 m/s. A particle size distribution (PSD) profile of MBR sludge showed a main peak at 40 m and a second peak at 0.1-1 m. The abundance of submicron particles at 2000 kDa was confirmed by a Liquid chromatography-Organic Carbon Detection (LC-OCD) analysis. The colloids responsible for the second peak in the PSD received high weighting factors (high filter cake formation potential) in the model optimization. In a lab-scale MBR, this critical crossflow velocity was between 0.75 and 1 m/s at 40 L/(m 2 h).
Application of conventional statistical monitoring methods to periodic processes can result in fr... more Application of conventional statistical monitoring methods to periodic processes can result in frequent false alarms and/or missed faults due to their common non-stationary behavior seen over a period. To address this, we propose to identify and use a stochastic statespace model that describes statistical behavior of the changes occurring from period to period. This model, when retooled as a periodically time-varying model, can be used for on-line monitoring and estimation with the aid of a Kalman filter. The same model can also be used for inferential estimation of the variables that ere difficult or slow to measure on-line. The proposed approach is applied to a simulation benchmark of waste-water treatment process, which exhibit strong diurnal changes in the feed stream, and compared against the Principal Component Analysis (PCA) and and Partial Least Squares (PLS) methods.
... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear m... more ... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear mapping )(⋅ Φ from an ... feature space. As a result, KPCA performs a nonlinear PCA in the input space (Romdhani et al., 1999). If ...
Batch processes lie at the heart of many industries; hence the effective monitoring and control o... more 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. #
DNA microarray technologies, which monitor simultaneously, the expression pattern of thousands of... more DNA microarray technologies, which monitor simultaneously, the expression pattern of thousands of individual genes in different biological systems have resulted in a tremendous increase of the amount of available gene expression data and have provided new insights into gene expression during development, within disease processes, and across species. However, microarray gene expression data are characterized by very high dimensionality (genes), relatively small numbers of samples (observations), irrelevant features, as well as collinear and multivariate characteristics. These features complicate the interpretation and analysis of microarray data, and the complexity of such data means that its analysis entails a high computational cost. This situation motivated the researchers to develop a new method for analyzing microarray data. In this paper, we propose a simple gene selection and multivariate fuzzy statistical analysis methods. The proposed method was applied to microarray data from leukemia patients; specifically, it was used to interpret the gene expression pattern and analyze the leukemia subtype whose expression profiles correlated with four cases of acute leukemia gene expression.
This article considers new and existing technologies for water reuse networks for water and waste... more This article considers new and existing technologies for water reuse networks for water and wastewater minimization. For the systematic design of water reuse networks, the theory of the water pinch methodology and the mathematical optimization are described, which are proved to be effective in identifying water reuse opportunities. As alternative solutions, evolutionary solutions and stochastic design approaches to water system design are also illustrated. And the project work flow and an example in a real plant are examined. Finally, as development is in the forefront in process industries, this paper will also explore some research challenges encountered in this field such as simultaneous water and energy minimization, energy-pinch design, and eco-industrial parks (EIP).
We propose a new algorithm for adaptive control and self tuning control, referred to as the gener... more We propose a new algorithm for adaptive control and self tuning control, referred to as the generalized damped least squares (GDLS) algorithm. This algorithm is constructed by adding a multi-step penalty for parameter variations to the objective function of the normal least squares algorithm to prevent the singularity problem that leads to estimation windup. We show that the proposed method has properties almost equivalent to those of the normal least squares method, which guarantees that the proposed algorithm is suitable for poorly excited situations. Simulation results show that the proposed method gives better estimation performance than previous methods in spite of its simplicity. The proposed method also shows good parameter tracking performance and no estimation windup. #
The goal of this work is the development of a suitable monitoring module, which is to be the firs... more The goal of this work is the development of a suitable monitoring module, which is to be the first module of an integrated fault detection and control system for the SHARON process. To model the process properly, different PCA models are tested. As a first step, PCA is used in an iterative manner to exclude data not considered to represent normal operational conditions and process behaviour from the original data set. To improve the performance of the identified model, it is decided to account for dynamics in the SHARON process by means of auto-regressive exogenous (ARX) structuring of data before the identification. A fruitful replacement of missing values for this purpose is done by means of a static PCA model. It is shown that the different criteria used in model selection lead to the same DPCA model. In this paper all steps of the monitoring module design are explained and the performance of different models is analyzed.
Most multivariate statistical monitoring methods based on principal component analysis (PCA) assu... more Most multivariate statistical monitoring methods based on principal component analysis (PCA) assume implicitly that the observations at one time are statistically independent of observations at past time and the latent variables follow a Gaussian distribution. However, in real chemical and biological processes, these assumptions are invalid because of their dynamic and nonlinear characteristics. Therefore, monitoring charts based on conventional PCA tend to show many false alarms and bad detectability. In this paper, a new statistical process monitoring method using dynamic independent component analysis (DICA) is proposed to overcome these disadvantages. ICA is a recently developed technique for revealing hidden factors that underlies sets of measurements followed on a non-Gaussian distribution. Its goal is to decompose a set of multivariate data into a base of statistically independent components without a loss of information. The proposed DICA monitoring method is applying ICA to the augmenting matrix with time-lagged variables. DICA can show more powerful monitoring performance in the case of a dynamic process since it can extract source signals which are independent of the auto-and cross-correlation of variables. It is applied to fault detection in both a simple multivariate dynamic process and the Tennessee Eastman process. The simulation results clearly show that the method e ectively detects faults in a multivariate dynamic process. ?
Batch processes are very important in most industries and are used to produce high-quality materi... more 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.
In this paper, a new nonlinear process monitoring technique based on kernel principal component a... more 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. ?
In this paper we propose a new statistical method for process monitoring that uses independent co... more In this paper we propose a new statistical method for process monitoring that uses independent component analysis (ICA). ICA is a recently developed method in which the goal is to decompose observed data into linear combinations of statistically independent components . Such a representation has been shown to capture the essential structure of the data in many applications, including signal separation and feature extraction. The basic idea of our approach is to use ICA to extract the essential independent components that drive a process and to combine them with process monitoring techniques. I 2 , I 2 e and SPE charts are proposed as on-line monitoring charts and contribution plots of these statistical quantities are also considered for fault identification. The proposed monitoring method was applied to fault detection and identification in both a simple multivariate process and the simulation benchmark of the biological wastewater treatment process, which is characterized by a variety of fault sources with non-Gaussian characteristics. The simulation results clearly show the power and advantages of ICA monitoring in comparison to PCA monitoring.
This study shows that an MBR pilot plant with UCT configuration is able to obtain high nutrient r... more This study shows that an MBR pilot plant with UCT configuration is able to obtain high nutrient removal efficiency already during start-up. The biological nutrient removal (BNR) efficiencies significantly increased towards the end of the experimental run, achieving a COD removal efficiency exceeding 94% and N removal efficiency in the range of 89 to 93%. P removal efficiencies in the range of 80 to 92% have been obtained. During the experimental period (4 months) the evolution of the activity of polyphosphate-accumulating organisms, obtained from P release and P uptake rates, showed a small increase in the activity of polyphosphateaccumulating organisms (PAOs) and denitrifying polyphosphate-accumulating organisms (DPAOs). The specific phosphate accumulation at the end of the experimental run amounted to 8.0 mg P g − 1 VSS h − 1 and 3.29 mg P g − 1 VSS h − 1 , for the PAOs and DPAOs respectively. Moreover, the DPAOs activity increased faster than PAOs activity, i.e. from 0.36 to 0.41 of phosphate uptake rate (PUR) ratio.
Application of conventional statistical monitoring methods to periodic processes can result in fr... more Application of conventional statistical monitoring methods to periodic processes can result in frequent false alarms and/or missed faults due to their common non-stationary behavior seen over a period. To address this, we propose to identify and use a stochastic statespace model that describes statistical behavior of the changes occurring from period to period. This model, when retooled as a periodically time-varying model, can be used for on-line monitoring and estimation with the aid of a Kalman filter. The same model can also be used for inferential estimation of the variables that ere difficult or slow to measure on-line. The proposed approach is applied to a simulation benchmark of waste-water treatment process, which exhibit strong diurnal changes in the feed stream, and compared against the Principal Component Analysis (PCA) and and Partial Least Squares (PLS) methods.
... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear m... more ... 1998). As shown in Fig. 1 (Romdhani et al., 1999), concep-tually, KPCA performs a nonlinear mapping )(⋅ Φ from an ... feature space. As a result, KPCA performs a nonlinear PCA in the input space (Romdhani et al., 1999). If ...
Batch processes lie at the heart of many industries; hence the effective monitoring and control o... more 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. #
Uploads
Papers by ChangKyoo Yoo