Multimode process monitoring based on Bayesian method

Multimode process monitoring has recently attracted much attention both in academy and industry. Conventional methods assume that either the process data are Gaussian in each operation mode, or some process knowledge should be incorporated, thus making the methods supervised. In this paper, a new unsupervised method is developed for multimode process monitoring, which is based on Bayesian inference and two‐step independent component analysis–principal component analysis (ICA–PCA) feature extraction strategy. ICA–PCA is first introduced for feature extraction and dimension reduction. By transferring the traditional monitoring statistic to fault probability in each operation mode, monitoring results in different operation modes can be easily combined by the Bayesian inference. Another contribution of the present paper is the development of a new fault identification method. Through analyses of the posterior probability and the joint probability for the monitored data sample, the correct operation mode or fault scenario can be identified. Three case studies are demonstrated to evaluate the feasibility and efficiency of the proposed method. Copyright © 2009 John Wiley & Sons, Ltd.     

Generalized Dice's coefficient‐based multi‐block principal component analysis with Bayesian inference for plant‐wide process monitoring

Plant‐wide process monitoring is challenging because of the complex relationships among numerous variables in modern industrial processes. The multi‐block process monitoring method is an efficient approach applied to plant‐wide processes. However, dividing the original space into subspaces remains an open issue. The loading matrix generated by principal component analysis (PCA) describes the correlation between original variables and extracted components and reveals the internal relations within the plant‐wide process. Thus, a multi‐block PCA method that constructs principal component (PC) sub‐blocks according to the generalized Dice coefficient of the loading matrix is proposed. The PCs corresponding to similar loading vectors are divided within the same sub‐block. Thus, the PCs in the same sub‐block share similar variational behavior for certain faults. This behavior improves the sensitivity of process monitoring in the sub‐block. A monitoring statistic T² corresponding to each sub‐block is produced and is integrated into the final probability index based on Bayesian inference. A corresponding contribution plot is also developed to identify the root cause. The superiority of the proposed method is demonstrated by two case studies: a numerical example and the Tennessee Eastman benchmark. Comparisons with other PCA‐based methods are also provided. Copyright © 2014 John Wiley & Sons, Ltd.     

Double‐step block division plant‐wide fault detection and diagnosis based on variable distributions and relevant features

Large‐scale process data in plant‐wide process monitoring are characterized by two features: complex distributions and complex relevance. This study proposes a double‐step block division plant‐wide process monitoring method based on variable distributions and relevant features to overcome this limitation. First, the data distribution is considered, and the normality test method called the D‐test is applied to classify the variables with the same distribution (i.e., Gaussian distribution or non‐Gaussian distribution) in a block. Thus, the second block division is implemented on both blocks obtained in the previous step. The mutual information shared between two variables is used to generate relevant matrixes of the Gaussian and non‐Gaussian blocks. The K‐means method clusters the vectors of the relevant matrix. Principal component analysis is conducted to monitor each Gaussian subblock, whereas independent component analysis is conducted to monitor each non‐Gaussian subblock. A composite statistic is eventually derived through Bayesian inference. The proposed method is applied to a numerical system and the Tennessee Eastman process data set. The monitoring performance shows the superiority of the proposed method. Copyright © 2015 John Wiley & Sons, Ltd.     

Bilinear modelling of batch processes. Part II: a comparison of PLS soft‐sensors

The on‐line monitoring of batch processes based on principal component analysis (PCA) has been widely studied. Nonetheless, researchers have not paid so much attention to the on‐line application of partial least squares (PLS). In this paper, the influence of several issues in the predictive power of a PLS model for the on‐line estimation of key variables in a batch process is studied. Some of the conclusions can help to better understand the capabilities of the proposals presented for on‐line PCA‐based monitoring. Issues like the convenience of batch‐wise or variable‐wise unfolding, the method for the imputation of future measurements and the use of several sub‐models are addressed. This is the first time that the adaptive hierarchical (or multi‐block) approach is extended to the PLS modelling. Also, the formulation of the so‐called trimmed scores regression (TSR), a powerful imputation method defined for PCA, is extended for its application with PLS modelling. Data from two processes, one simulated and one real, are used to illustrate the results. Copyright © 2008 John Wiley & Sons, Ltd.     

Quantification and methodology issues in multivariate analysis of ToF‐SIMS data for mixed organic systems

Multivariate methods, such as principal component analysis (PCA) and multivariate curve resolution (MCR), are often employed to aid the analysis of large complex data sets such as time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS) images. There is, however, much confusion over the most appropriate choice of method for any given application and the effects of data preprocessing, which is exacerbated by the confusing terminologies and the use of jargon in this field. In the present study, a simple model system consisting of a ToF‐SIMS image of an immiscible polymer blend is used to evaluate PCA and MCR in the accurate identification, localisation and quantification of the phase‐separated polymer domains, using four data preprocessing methods (no scaling, normalisation, variance scaling and Poisson scaling). This highlights significant issues and challenges in the quantitative multivariate analysis of mixed organic systems, including the discrimination of chemically significant features from experimental noise, the resolution of weak chemical contributions and potential bias introduced by data preprocessing. Multivariate analysis using Poisson scaling, identified as the most suitable data preprocessing method for both PCA and MCR, demonstrates a marked improvement upon traditional (manual) analysis and provides valuable additional information that is difficult to detect using traditional analysis. Using these results, we present recommendations for the optimum use of multivariate analysis by analysts and provide guidance on selecting the most appropriate methods. Confusing terminology is also clarified. © Crown copyright 2008. Reproduced with the permission of Her Majesty's Stationery Office. Published by John Wiley & Sons, Ltd.     

Fault diagnosis using kNN reconstruction on MRI variables

In industrial processes, investigating the root causes of abnormal events is a crucial task when process faults are detected; isolating the faulty variables provides additional information for investigating the root causes of the faults. The traditional contribution plot is a popular and perspicuous tool to isolate faulty variables. However, this method can only determine one faulty variable (the biggest contributor) when there are several variables out of control at the same time. In the presented work, a novel fault diagnosis method is derived using k‐nearest neighbor (kNN) reconstruction on maximize reduce index (MRI) sensors; it is aimed at identifying all fault variables precisely. This method can identify the faulty variables effectively through reconstructing MRI variables one by one. A numerical example focuses on validating the performance of kNN missing data analysis method firstly, then multi‐sensors fault identification results are also given. Tennessee Eastman process is provided to demonstrate that the proposed approach can identify the responsible variables for the multiple sensors fault. Copyright © 2015 John Wiley & Sons, Ltd.     

Geographical classification of Epimedium based on HPLC fingerprint analysis combined with multi‐ingredients quantitative analysis

A reliable and comprehensive method for identifying the origin and assessing the quality of Epimedium has been developed. The method is based on analysis of HPLC fingerprints, combined with similarity analysis, hierarchical cluster analysis (HCA), principal component analysis (PCA) and multi‐ingredient quantitative analysis. Nineteen batches of Epimedium , collected from different areas in the western regions of China, were used to establish the fingerprints and 18 peaks were selected for the analysis. Similarity analysis, HCA and PCA all classified the 19 areas into three groups. Simultaneous quantification of the five major bioactive ingredients in the Epimedium samples was also carried out to confirm the consistency of the quality tests. These methods were successfully used to identify the geographical origin of the Epimedium samples and to evaluate their quality.     

A real‐time Mooney‐viscosity prediction model of the mixed rubber based on the Independent Component Regression‐Gaussian Process algorithm

With the rapid development of rubber industry, it becomes more and more important to improve the performance of the quality control system of rubber mixing process. Unfortunately, the large measurement time delay of Mooney viscosity, one of the most important quality parameters of mixed rubber, badly blocks the further development of the issue. The independent component regression‐Gaussian process (ICR‐GP) algorithm is used to solve such typical nonlinear “black‐box” regression problem for the first time to predict Mooney viscosity. In the ICR‐GP method, the non‐Gaussian information is extracted by the independent component regression method firstly, and then the residual Gaussian information is extracted by the Gaussian process method. Meanwhile, both the linear and nonlinear relationships between the input and output variables can be extracted through the ICR‐GP method. With the fact that there is no need to optimize parameters, the ICR‐GP method is especially suitable for “black‐box” regression problems. The highest prediction accuracy was achieved at M = 0.8765 (the root mean square error), which was high enough considering the measuring accuracy (M = ±0.5) of the Mooney viscometer. It is by using the online‐measured rheological parameters as the input variables that the measurement time delay of Mooney viscosity could be dramatically decreased from about 240 to 2 min. Consequently, such Mooney‐viscosity prediction model is very helpful for the development of the rubber mixing process, especially of the emerging one‐step rubber mixing technique. The practical applications performed on the rubber mixing process in a large‐scale tire factory strongly proved the outstanding regression performance of this ICR‐GP Mooney‐viscosity prediction model. Copyright © 2012 John Wiley & Sons, Ltd.     

Using Unfold‐PCA for batch‐to‐batch start‐up process understanding and steady‐state identification in a sequencing batch reactor

In chemical and biochemical processes, steady‐state models are widely used for process assessment, control and optimisation. In these models, parameter adjustment requires data collected under nearly steady‐state conditions. Several approaches have been developed for steady‐state identification (SSID) in continuous processes, but no attempt has been made to adapt them to the singularities of batch processes. The main aim of this paper is to propose an automated method based on batch‐wise unfolding of the three‐way batch process data followed by a principal component analysis (Unfold‐PCA) in combination with the methodology of Brown and Rhinehart 2 for SSID. A second goal of this paper is to illustrate how by using Unfold‐PCA, process understanding can be gained from the batch‐to‐batch start‐ups and transitions data analysis. The potential of the proposed methodology is illustrated using historical data from a laboratory‐scale sequencing batch reactor (SBR) operated for enhanced biological phosphorus removal (EBPR). The results demonstrate that the proposed approach can be efficiently used to detect when the batches reach the steady‐state condition, to interpret the overall batch‐to‐batch process evolution and also to isolate the causes of changes between batches using contribution plots. Copyright © 2007 John Wiley & Sons, Ltd.     

Simultaneous determination of three anticon‐vulsants using hydrophilic interaction LC‐MS

A specific and automated method was developed to quantify the anticonvulsants gabapentin, pregabalin and vigabatrin simultaneously in human serum. Samples were prepared with a protein precipitation. The hydrophilic interaction chromatography (HILIC) with a mobile phase gradient was used to divide off ions of the matrix and for separation of the analytes. Four different HILIC‐columns and two different column temperatures were tested. The Tosoh‐Amid column gave the best results: single small peaks. The anticonvulsants were detected in the multiple reaction monitoring mode (MRM) with ESI‐MS‐MS. Using a volume of 100 μL biological sample the lowest point of the standard curve, i.e. the lower LOQs were 312 ng/mL. The described HILIC‐MS‐MS method is suitable for therapeutic drug monitoring and for clinical and pharmcokinetical investigations of the anticonvulsives.     

Multi‐Channel Piezoelectric Quartz Crystal Sensor with Principal Component Analysis and Back‐Propagation Neural Network for Organic Pollutants from Petrochemical Plants

A multi‐channel piezoelectric quartz crystal gas sensor comprising arrays coated with various organic materials and a home‐made computer interface for data processing were prepared and employed to detect six kinds of common organic pollutants from petrochemical plants including benzene, styrene, chloroform, octane, hexene and hexyne. The principal component analysis (PCA) method was employed to select six kinds of appropriate coating materials for these organic pollutants from 22 adsorbents onto piezoelectric crystals. After performing a PCA assay, six representative coating materials, namely Polyisobutylene, Poly(dimethylsiloxane) (SE30), 4‐tert‐Butylcalix[6]arene, Cholesteryl chloroformate, C60‐Polyphenyl acetylene (C60‐PPA) and Ag(I)/cryptand‐2,2/Ethylene diamine/NH₃/Polyvinyl chloride were selected. Moreover, effects of coating load of adsorbents and concentration of pollutants were also investigated. Three kinds of recognition techniques including 2D PCA score map, radar plot and back‐propagation neural network (BPN) were employed for qualitative analysis of these organic pollutants, and a quantitative analysis method could be established by creating calibration curves for each organic pollutant. This homemade multi‐channel piezoelectric quartz crystal gas sensor showed a good detection limit of 0.068‐1.127 mg/L for these organic pollutants. The multi‐channel piezoelectric gas sensor exhibited good reproducibility with a relative standard deviation (RSD) of 1.1‐9.6%. Furthermore, this multi‐channel piezoelectric crystal detection system with BPN recognition technique was also utilized to successfully distinguish and identify each component of the mixture of organic gas samples. Multivariate linear regression (MLR) analysis was employed to quantitatively compute the concentration of species in the organic mixtures.     

A Bayesian sparse reconstruction method for fault detection and isolation

This article develops a Bayesian method for fault detection and isolation using a sparse reconstruction framework. The normal/training data is assumed to follow a signal‐plus‐noise model, and an indicator matrix is used to show whether the test data is from a faulty process. The distribution of the indicator matrix is modeled by a Laplacian distribution, which forces the indicator matrix to be a sparse one, and a Gibbs sampler is derived to obtain the estimation/reconstruction of the indicator matrix, the unobserved signals, and other parameters like signal mean, covariance, and noise variance. The faulty variables can then be detected and isolated by inspecting whether corresponding rows of the indicator matrix are zero. The proposed Bayesian approach is data driven; it allows for simultaneous fault detection and isolation. A simulation study and an industrial case study are used to test the performance of the proposed method. Copyright © 2015 John Wiley & Sons, Ltd.     

Purification of alkaloids from Corydalis yanhusuo W. T. Wang using preparative 2‐D HPLC

Two‐dimensional preparative multi‐channel parallel high performance liquid chromatography was successfully applied for the first time to isolate and purify alkaloids from Corydalis yanhusuo. The experiments were performed in off‐line mode using the same preparative chromatographic column with pH 3.5 in the first and pH 10.0 in the second separation dimension. In the preparative process, UV‐triggered fraction collection was used in the first dimension while UV and MS‐triggered collection were used in the second dimension for reasons of sensitivity and complementarity. Two pure compounds and nine fractions were obtained in the first dimension. Then two representative fractions were further purified in the second dimension and six pure compounds were obtained. The results demonstrated that this procedure is an effective approach for the preparative isolation and purification of alkaloids from Corydalis yanhusuo. Based on the different pH values of the mobile phase in this method, it is also suitable for the preparative isolation and purification of other compounds from TCMs which are sensitive to the pH of the solutions. Moreover, this method will be a promising tool for the purification of low content compounds from natural products.     

Multi‐Channel Surface Acoustic Wave Sensors Based on Principal Component Analysis (PCA) and Linear Discriminate Analysis (LDA) for Organic Vapors

A multi‐channel surface acoustic wave (SAW) detection system which is employed to detect various organic molecules in a static system was prepared using 315 MHz one‐port quartz resonators and a home‐made computer interface for signal acquisition and data process. The oscillating frequency of the quartz crystal decreases on adsorption of organic molecules on the coating materials. The principal component analysis (PCA) method with SAS software was applied to select the appropriate coating materials onto the SAW crystals for organic vapors, e.g. hexane, 1‐hexene, 1‐hexyne, 1‐propanol, propionaldehyde, propionic acid, and 1‐propylamine. A dataset for a multi‐channel sensor with 19 SAW crystals for 7 analyses was collected after comparing the correlation between the 19 coating materials and the first six principal component (PC) factor. Furthermore, linear discriminate analysis (LDA) with SPSS software and a profile discrimination map were also applied and discussed for the discrimination of these organic vapors. These organic molecules could be clearly distinguished by the six‐channel SAW static sensor. The effect of concentration for various organic vapors was investigated and discussed.     

Adsorption of genetically engineered proteins studied by time‐of‐flight secondary ion mass spectrometry (TOF‐SIMS). Part B: hierarchical cluster analysis (HCA)

In Part A, we adopted principal component analysis (PCA) for the analysis of TOF‐SIMS data to assess the binding specificity of GBP‐1 to metallic Au, Ag and Pd. Within a given set of data, PCA aids in the interpretation of the TOF‐SIMS spectra by capitalizing on the differences from one spectrum to another. In Part B, we introduce another multivariate statistical method called ‘hierarchical cluster analysis (HCA)’, where visualization of the similarity and difference in data is readily observed, from which a variety of adsorption conditions of GBP‐1 were characterized. Copyright © 2007 John Wiley & Sons, Ltd.     

Intermolecular Carbonyl–olefin Metathesis with Vinyl Ethers Catalyzed by Homogeneous and Solid Acids in Flow

The carbonyl–olefin metathesis reaction has experienced significant advances in the last seven years with new catalysts and reaction protocols. However, most of these procedures involve soluble catalysts for intramolecular reactions in batch. Herein, we show that recoverable, inexpensive, easy to handle, non‐toxic, and widely available simple solid acids, such as the aluminosilicate montmorillonite, can catalyze the intermolecular carbonyl–olefin metathesis of aromatic ketones and aldehydes with vinyl ethers in‐flow, to give alkenes with complete trans stereoselectivity on multi‐gram scale and high yields. Experimental and computational data support a mechanism based on a carbocation‐induced Grob fragmentation. These results open the way for the industrial implementation of carbonyl–olefin metathesis over solid catalysts in continuous mode, which is still the origin and main application of the parent alkene–alkene cross‐metathesis.     

At‐line multi‐angle light scattering detector for faster process development in enveloped virus‐like particle purification

At‐line static light scattering and fluorescence monitoring allows direct in‐process tracking of fluorescent virus‐like particles. We have demonstrated this by coupling at‐line multi‐angle light scattering and fluorescence detectors to the downstream processing of enveloped virus‐like particles. Since light scattering intensity is directly proportional to particle concentration, our strategy allowed a swift identification of product containing fractions and rapid process development. Virus‐like particles containing the Human Immunodeficiency Virus‐1 Gag protein fused to the Green Fluorescence protein were produced in Human Embryonic Kidney 293 cells by transient transfection. A single‐column anion‐exchange chromatography method was used for direct capture and purification. The majority of host‐cell protein impurities passed through the column without binding. Virus‐like particles bound to the column were eluted by linear or step salt gradients. Particles recovered in the step gradient purification were characterized by nanoparticle tracking analysis, size exclusion chromatography coupled to multi‐angle light scattering and fluorescence detectors and transmission electron microscopy. A total recovery of 66% for the fluorescent particles was obtained with a 50% yield in the main product peak. Virus‐like particles were concentrated 17‐fold to final a concentration of 4.45 × 10¹⁰ particles/mL. Simple buffers and operation make this process suitable for large scale purposes.     

New analysis of the optimization of electromagnetic shielding properties using conducting polymers and a multi‐objective approach

Recently, an approach was proposed to optimize multi‐layer shields of polyaniline–polyurethane (PAni/PU) conducting composites in the microwave band. Though by this method shields for different applications can be obtained which are light‐weight and offer a low percolation threshold, the full potential of the design process could not be tapped since the underlying optimization problem includes only one objective. In this work we go one step beyond and re‐formulate the design problem as a multi‐objective optimization problem (MOP). To be more precise, we involve simultaneously the shielding efficiency as well as the weight and the cost of the material—i.e. all the requirements for modern shielding materials—within the optimization process. After having stated the model we present two possible ways to approximate the solution set—the so‐called Pareto set—and address the related and important decision‐making problem. All steps are demonstrated on a particular three‐layered composite in order to show the applicability of the novel approach. Copyright © 2008 John Wiley & Sons, Ltd.     

Synthesis of Co(II), Cu(II), Hg(II), UO2(II) and Pb(II) binuclear nanometric complexes from multi‐donor ligand: Spectral,modeling, quantitative structure–activity relationship,docking and antitumor studies

New synthesis of Co(II), Cu(II), Hg(II), UO₂(II) and Pb(II) binuclear nanometric complexes derived from multi‐donor ligand is reported. Structural and molecular formulae of all isolated complexes were established. Bi‐negative hexa‐dentate mode of ligand was proposed for the two central atoms in all complexes. Infrared, UV–visible, magnetic moments, electron paramagnetic resonance, thermogravimetric analysis and elemental analysis were used to build all structural formulae. X‐ray diffraction and scanning electron microscopy were used to determine the morphological features of the compounds and to compute particle sizes. Theoretical computations were implemented to support the proposed formulae. Kinetic study was executed over suitable stages to clarify the comparative stabilities. The DFT/B3LYP method, using the Gaussian 09 program, was utilized for optimizing the distribution of atoms over all compounds except the UO₂²⁺ complex. This exclusion refers to an inability to find a suitable method. Significant parameters were estimated using frontier energies (highest occupied and lowest unoccupied molecular orbitals) and data log file. A quantitative structure–activity relationship study applying HyperChem was executed for the organic compound tautomer forms to give a significant view about their biological character. AutoDock tools 4.2 were used to investigate the biological trend of organic compounds (keto and enol) against three types of proteins. The types were chosen related to in vitro investigation: breast, prostate and liver carcinoma proteins. IC₅₀ values indicated insignificant inhibition activity towards all carcinoma cell lines under investigation, except for the Hg(II) complex which displayed distinct activity against breast carcinoma compared with reference drug (doxorubicin).     

Determination of rank by median absolute deviation (DRMAD): a simple method for determining the number of principal factors responsible for a data matrix

Median absolute deviation (MAD) is a well‐established statistical method for determining outliers. This simple statistic can be used to determine the number of principal factors responsible for a data matrix by direct application to the residual standard deviation (RSD) obtained from principal component analysis (PCA). Unlike many other popular methods the proposed method, called determination of rank by MAD (DRMAD), does not involve the use of pseudo degrees of freedom, pseudo F‐tests, extensive calibration tables, time‐consuming iterations, nor empirical procedures. The method does not require strict adherence to normal distributions of experimental uncertainties. The computations are direct, simple to use and extremely fast, ideally suitable for online data processing. The results obtained using various sets of chemical data previously reported in the chemical literature agree with the early work. Limitations of the method, determined from model data, are discussed. An algorithm, written in MATLAB format, is presented in the Appendix. Copyright © 2008 John Wiley & Sons, Ltd.