基于PLS-LSSVM的谷氨酸发酵产物浓度预测建模

针对谷氨酸发酵过程关键生化参数难以在线检测给发酵优化控制带来困难问题,基于谷氨酸5 L发酵罐发酵过程,建立基于偏最小二乘（PLS）和最小二乘向量机（LSSVM）相结合的谷氨酸浓度预测模型;利用PLS对输入变量进行特征提取降低维数和消除相关性,以简化模型和提高模型精度。为确定谷氨酸发酵最佳预测模型,简化后的预测模型与发酵动力学模型进行比较;实验结果表明,简化后的耦合模拟退火（coupled simulated annealing,CSA）对参数进行优化的LSSVM模型具有最好预测性能,相对PLS预测模型和发酵动力学模型具有明显优势,均方根误差分别为1.597、8.49和2.934,可以为谷氨酸发酵过程操作及时调整及优化控制提供有效指导。     

Development of calibration models for estimation of monomer concentration by Raman spectroscopy during emulsion polymerization: Facing the medium heterogeneity

This work compares different calibration models for the estimation of monomer concentrations by Raman spectroscopy during semicontinuous emulsion copolymerization reactions. The limitations of these models are discussed in terms of a complex reaction, namely the copolymerization of vinyl acetate and butyl acrylate, whose monomers present overlapping Raman spectra, especially the C=C stretching band. Additionally, the copolymerization was monitored in a spectroscopic setup arranged for fast spectral acquisition, which resulted in a low signal‐to‐noise ratio. These realistic conditions for in‐line monitoring of emulsion copolymerization, i.e., considerable noise level in the spectra and medium heterogeneity, are discussed in the context of different approaches for adjusting the calibration model and the ensuing model limitations. It was verified that combining data obtained during reactions with synthetic samples is interesting from the statistical point of view, since in this way it is possible to produce data sets with a wide range of variation, allowing the accurate estimation of statistical parameters. These parameters are of major importance for process variables and product property estimations, especially if they are to be used for process control and decision making purposes. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 93: 1136–1150, 2004     

Optimal selection of time resolution for batch data analysis. Part I: Predictive modeling

Soft sensors based on multiway partial least squares (MW‐PLS) are often used to estimate, in useful time, the end quality of batch processes, due to their ability to deal with high dimensional and noisy data. However, PLS and its variants only bring parsimony to the variables' mode. The time mode, which is the main source of complexity in MW‐PLS, remains unchanged. Parsimony on the time dimension can be achieved by manipulating the variables' resolution or granularity. In this article, we address the optimal selection of resolution for each individual batch variable, as an additional degree of freedom for maximizing the predictive performance of industrial soft sensors. The proposed methodology will conduct, simultaneously, the optimal selection of (1) variables, (2) resolutions, and (3) stages. At the end, a multiresolution PLS model (MR‐PLS) will be obtained, that optimally predicts the batch‐end quality within the class of all MW‐PLS approaches. © 2018 American Institute of Chemical Engineers AIChE J, 64: 3923–3933, 2018     

A Bayesian framework for real‐time identification of locally weighted partial least squares

Just‐in‐time (JIT) learning methods are widely used in dealing with nonlinear and multimode behavior of industrial processes. The locally weighted partial least squares (LW‐PLS) method is among the most commonly used JIT methods. The performance of LW‐PLS model depends on parameters of the similarity function as well as the structure and parameters of the local PLS model. However, the regular LW‐PLS algorithm assumes that the parameters of the similarity function and structure of the local PLS model are known and do not fully utilize available knowledge to estimate the model parameters. A Bayesian framework is proposed to provide a systematic way for real‐time parameterization of the similarity function, selection of the local PLS model structure, and estimation of the corresponding model parameters. By applying the Bayes' theorem, the proposed framework incorporates the prior knowledge into the identification process and takes into account the different contribution of measurement noises. Furthermore, Bayesian model structure selection can automatically deal with the model complexity problem to avoid the overfitting issue. The advantages of this new approach are highlighted through two case studies based on the real‐world near infrared data. © 2014 American Institute of Chemical Engineers AIChE J, 61: 518–529, 2015     

露天矿卡车实时监控调度系统的研制

介绍了一个露天矿卡车实时监控调度系统的研制情况。研制的内容主要包括了系统功能及总体设计、调度数据库设计、车流规划软件设计、实时调度软件设计，着重介绍了系统作业方式设计。该系统成功地进行了室内和现场试验，取得了满意的结果。系统的应用可以减少铲、车的相互等待时间，提高生产效率。     

Direct sampling of orujo oil for determining residual hexane by using a chemsensor

Hexane is used to extract edible oils from oleaginous seeds. The detection of hexane in orujo oil is mandatory, as its presence in the final product may negatively affect human health. Headspace-GC is the technique of choice for determining residual solvent in foods. In the present work, a new instrument based on the headspace principle and mass spectrometric detection without chromatographic separation, ChemSensor, is proposed for the direct screening of orujo oil to determine residual hexane. This instrument provided an overall response, corresponding to the volatiles profile, including that of hexane, which could not be directly discriminated. By selecting the m/z values corresponding to n-hexane (major component of commercial hexane), the selectivity of the method was good enough to determine residual hexane in the range of 2.0–65 μg mL⁻¹ (corresponding to 2.3–75.6 mg of hexane per kg of oil) with high precision. The detection limit achieved (0.7 mg per kg of oil) was lower than the maximum residual limit established by the European Union (5 mg per kg of oil). Two multivariate techniques, partial least squares and principal components regression (PCR), were compared with univariate regression; PCR provided the best results.     

Application of MIA for a QSAR Study of Inhibitory Activity of DHP Derivatives and Design of New Compounds Using WT and GA for Pixel Processing

A quantitative structure-activity relationship (QSAR) modeling was carried out for the prediction of inhibitory activity of dihydropyridine (DHP) derivatives known as calcium channel blocker (CCB) drugs. Partial least squares (PLS) algorithm was used for prediction of inhibitory activity of calcium channel antagonists as a function of the bidimensional images. In the present study, it is investigated that the effect of pixel selection by application of genetic algorithms (GAs) for PLS model, because of the GAs is very useful in the variable selection in modeling. Pre-processing methods such as wavelet transform (WT) were also used to enhance the predictive power of multivariate calibration methods. The subset of pixels, which resulted in the low prediction error, was selected by GA. To evaluate the models applied in this study (PLS, GA-PLS and WT-GA-PLS), the inhibitory activities of several compounds, not included in the modeling procedure, were predicted. The results of models showed high prediction ability with root mean square error of prediction (RMSEP) of 0.51, 0.39 and 0.17 for PLS, GA-PLS and WT-GA-PLS, respectively. The WT-GA-PLS method was employed to predict the inhibitory activity of the new antagonists.     

Multivariate model for the prediction of total phenolic acids in crude extracts of polyphenols from canola and rapessed meals: A preliminary study

The feasibility of using UV spectrophotometry to develop multivariate models for prediction of total phenolic acids content in crude polyphenol extracts from defatted canola and rapeseed meals was investigated. The polyphenols were extracted from the meals with methanol/acetone/water (7∶7∶6, by vol). Partial least squares regression was used to correlate the spectral data of the crude polyphenols in methanol between 320 and 355 nm with the total phenolic acid content in canola and rapeseed meals. The Folin-Denis assay was used to provide reference data for creating the model. The predictive ability of the model is good, as indicated by the RPD value (the ratio of the SD of data to the standard error of calibration) of 3.84.     

The use of the F.p.l.c.R system in method development and process monitoring for industrial protein chromatography

The f.p.l.c.^R for high performance biomolecular purification is now widely used in laboratory and pilot scale separations, both as a means of developing new methods for purification and for analysing the purity of intermediary and final products. The speed of the technique means that it is possible to analyse fractions from a process step and to receive the results within 15–30 min. Two aspects are examined in this paper: the monitoring of protein product formation during fermentation and the analysis of fractions in large scale chromatography. The suitability of the technique in an operating process is discussed.     

Multivariate calibration of fourier transform infrared spectra for determining thiobarbituric acid-reactive substance content in palm oil

Fourier transform infrared (FTIR) spectra of palm oil samples between 2900 and 2800 cm⁻¹ and 1800 and 1600 cm⁻¹ were used to compare different multivariate calibration techniques for quantitative determination of their thiobarbituric acid-reactive substance (TBARS) content. Fifty spectra (in duplicate) of palm oil with TBARS values between 0 and 0.25 were used to calibrate models based on partial least squares (PLS) and principal components regression (PCR) analyses with different baselines. The methods were compared for the number of factors, coefficients of determination (R ²), and accuracy of estimation. The standard errors of prediction (SEP) were calculated to compare their predictive ability. The calibrated models generated three to eight factors, R ² of 0.9414 to 0.9803, standard error of estimation (SEE) of 0.0063 to 0.0680, and SEP of 1.20 to 6.67.     

A novel multiple linear multivariate NIR calibration model‐based strategy for in‐line monitoring of continuous mixing

The capability of near infra‐red (NIR) spectroscopy to predict many different variables, such as concentration and humidity, has been demonstrated in many published works. Several of those articles have been in the subject of real time prediction of continuous operations. However, those demonstrations have been for narrow ranges of the variables, especially for powder concentration, which could present a nonlinear behavior of the NIR absorbance as a function of the entire range of concentration. This work developed a novel strategy to predict the entire range of powder concentration using multiple linear NIR calibration models. The root mean standard error of prediction and relative standard deviation (RSD) parameters were used to establish the number of the multiple linear calibration models; other statistical features were used to establish the correct prediction. It was found that a minimum number of linear partial least squares (PLS) calibration models were necessary to accurately predict the range from 0 to 100% w/w. This technique could also be used with other nonlinear behaviors. © 2014 American Institute of Chemical Engineers AIChE J, 60: 3123–3132, 2014     

Development of a Measuring Method for the Determination of Bisulfite and Sulfite in Seawater

In conventional reverse osmosis processes for seawater desalination, a disinfection of the process stream with chlorine compounds is carried out for antifouling. After disinfection the reduction agent Na₂S₂O₅ is used for the removal of residual chlorine in a strongly overstoichiometric way in order to protect the membranes from oxidational damages. To save chemicals a controlled dosing of Na₂S₂O₅ based on a reliable concentration measurement is desirable. Therefore, a measuring method for the determination of the sulfur(IV) components bisulfite and sulfite in seawater is developed based on the combination of UV spectroscopy and a PLS regression method. Experimental results as well as the development of the regression model for sulfur species in ultrapure and seawater is described.     

Development of calibration models for quality control in the production of ethylene/propylene copolymers by FTIR spectroscopy,multivariate statistical tools,and artificial neural networks

Principal component regression (PCR), partial least squares (PLS), StepWise ordinary least squares regression (OLS), and back‐propagation artificial neural network (BP‐ANN) are applied here for the determination of the propylene concentration of a set of 83 production samples of ethylene–propylene copolymers from their infrared spectra. The set of available samples was split into (a) a training set, for models calculation; (b) a test set, for selecting the correct number of latent variables in PCR and PLS and the end point of the training phase of BP‐ANN; (c) a production set, for evaluating the predictive ability of the models. The predictive ability of the models is thus evaluated by genuine predictions. The model obtained by StepWise OLS turned out to be the best one, both in fitting and prediction. The study of the breakdown number of samples to be included in the training set showed that at least 52 experiments are necessary to build a reliable and predictive calibration model. It can be concluded that FTIR spectroscopy and OLS can be properly employed for monitoring the synthesis or the final product of ethylene–propylene copolymers, by predicting the concentration of propylene directly along the process line. © 2008 Wiley Periodicals, Inc. J Appl Polym Sci, 2008     

A flow injection analysis system for on‐line monitoring of cutinase activity at outlet of an expanded bed adsorption column almost in real time

Expanded bed adsorption (EBA) was used to recover, concentrate and purify Fusarium solani pisi cutinase, secreted by a recombinant Saccharomyces cerevisiae strain, directly from a whole fermentation culture. A flow injection analysis (FIA) system for monitoring Fusarium solani pisi cutinase based on microencapsulation of p‐nitrophenylbutyrate (p‐NPB) in a micellar system (sodium cholate, tetrahydrofuran, phosphate buffer) was developed for monitoring this target enzyme at the outlet tube of the EBA column. Slight differences in yeast cultivation conditions during cutinase production may influence the fermentation performance, which affects directly the adsorption of cutinase during the loading step and consequently the efficiency of the EBA process. This effect can be especially relevant when it is necessary to stop the application of feedstock to the EBA column when the outlet concentration (A) of the desired product is lower than 5% of the feed concentration (A_o). Excellent correlations between the FIA system and the off‐line analytical method for monitoring cutinase activity during the different EBA steps were obtained. Additionally, the blocking/fouling of the sample injector and tubes of the FIA system initially observed were eliminated due to the excellent surfactant properties of the sodium cholate contained in the phosphate buffer and used to dilute the enzyme samples. This FIA system was shown to be a powerful analytical tool for monitoring cutinase activity almost in real time (45–60 s), maximizing enzyme adsorption while minimizing product loss and consequently maximizing the recovery yield of the product. Copyright © 2006 Society of Chemical Industry     

Mid‐infrared spectroscopy as a tool for real‐time monitoring of ethanol absorption in glycols

Fast, simple, accurate, and inexpensive methods for obtaining analyte concentration data are desirable in the industrial sector. In the present study, the use of Fourier transform mid‐infrared (FT‐MIR) spectroscopy, combined with partial least squares (PLS) regression, was investigated as a tool for real‐time monitoring of processes of ethanol absorption in glycols. Calibration was performed using simple synthetic samples containing ethanol, water, and monoethylene glycol (MEG) or diethylene glycol (DEG). The PLS models presented excellent performance, with correlation coefficients (R²) close to unity and root‐mean‐square errors of cross‐validation (RMSECV) and prediction (RMSEP) lower than 2% of the calibration data ranges for both analytes (ethanol and water) in both absorbents (MEG and DEG). The monitoring technique developed has potential to be applied in absorption processes and could also be used in other large‐scale unit operations, providing information in real time and enhancing process control.