Control of a Batch Polymerization System Using Hybrid Neural Network ‐ First Principle Model

In this work, the utilization of neural network in hybrid with first principle models for modelling and control of a batch polymerization process was investigated. Following the steps of the methodology, hybrid neural network (HNN) forward models and HNN inverse model of the process were first developed and then the performance of the model in direct inverse control strategy and internal model control (IMC) strategy was investigated. For comparison purposes, the performance of conventional neural network and PID controller in control was compared with the proposed HNN. The results show that HNN is able to control perfectly for both set points tracking and disturbance rejection studies.     

Simulation Studies on Growth and Death of Microorganisms Using the Oil‐Degrading Bacteria Petrotoga sp.

Three models were developed to estimate the potential of the selected bacteria Petrotoga sp., a thermophilic anaerobic oil‐degrading microorganism. Fourteen data sets of these bacteria were simulated by a multilayer feed‐forward neural network and an adaptive neuro‐fuzzy interference system. Twelve data sets served for training and two for testing these models. A simplified numerical model was performed assuming two phases in the growth process of oil‐degrading microorganisms, the logarithmic growth phase and the death phase. Comparison between these models in predicting bacterial cell concentration for different data sets indicates little difference between the overall average relative errors of the three methods and that all can be applied for prediction. Effects of salinity concentration, amount of yeast extract, and temperature on bacterial cell concentration were simulated by numerical and neural network models.     

Hybrid modeling of inulinase bio‐production process

BACKGROUND: A potential application of inulinase in the food industry is the production of fructooligosaccharides (FOS) through transfructosilation of sucrose. Besides their ability to increase the shelf‐life and flavor of many products, FOS have many interesting functional properties. The use of an industrial medium may represent a good, cost‐effective alternative to produce inulinase, since the activity of the enzyme produced may be improved or at least remain the same compared with that obtained using a synthetic medium. Thus, inulinase production for use in FOS synthesis is of considerable scientific and technological appeal, as is the development of a reliable mathematical model of the process. This paper describes a hybrid neural network approach to model inulinase production in a batch bioreactor using agroindustrial residues as substrate. The hybrid modeling makes use of a series artificial neural network to estimate the kinetic parameters of the process and the mass balance as constitutive equations. RESULTS: The proposed model was shown to be capable of describing the complex behavior of inulinase production employing agroindustrial residues as substrate, so that the mathematical framework developed is a useful tool for simulation of this process. CONCLUSION: The hybrid neural network model developed was shown to be an interesting alternative to estimate model parameters since complete elucidation of the phenomena and mechanisms involved in the fermentation is not required owing to the black‐box nature of the ANN used as parameter estimator. Copyright © 2010 Society of Chemical Industry     

Multi‐Objective Optimization of Pseudo‐Dynamic Operation of Naphtha Pyrolysis by a Surrogate Model

A simple pseudo‐dynamic surrogate model is developed in the framework of the state space model with the feed‐forward neural network to replace the complex free radical pyrolysis model. The surrogate model is then applied to investigate the multi‐objective optimization of two key performance objectives with distinct contradiction: the mean yields of key products and the day mean profits. The ?‐constraint method is employed to solve the multi‐objective optimization problem, which provides a broad range of operation conditions depicting tradeoffs of both key objectives. The Pareto‐optimal frontier is successfully obtained and five selected cases on the frontier are discussed, suggesting that flexible operations can be performed based on industrial demands.     

Development of a hybrid neural network model to predict feeding method in fed‐batch cultivation for enhanced recombinant streptokinase productivity in Escherichia coli

BACKGROUND: A simple and efficient model for enhancing production of recombinant proteins is essential for cost effective development of processes at industrial scale. A hybrid neural network (HNN) model is proposed combining an unstructured model and neural network to predict the feeding method for the post‐induction phase of fed‐batch cultivation for increased recombinant streptokinase activity in Escherichia coli. RESULTS: The parameters of the unstructured model were estimated from experiments conducted with various feeding methods. The simulated model described the dynamics of the process satisfactorily, however, its predictive capability of the process for different feeding methods is limited due to wide disparity in process parameters. In contrast, a neural network model trained to map the variations in process parameters to state variables complements the ‘first principle’ model in predicting the state variables effectively. CONCLUSIONS: The HNN model is able to predict the product profile for different substrate feed rates. Further, the average volumetric streptokinase activity predicted by the HNN model matches closely the experimental values for fed‐batches having high as well as low streptokinase activity. The HNN model developed in this study could facilitate development of a process for recombinant protein production with minimum number of experiments. Copyright © 2011 Society of Chemical Industry     

乙烯裂解炉反应与传热耦合的智能混合建模与模拟

乙烯裂解炉辐射段的过程模拟,一般由管内反应过程模型与管外传热模型两部分组成,具有非线性、强耦合的特点。其中管外传热模型涉及变量参数众多、求解过程耗时长。针对这一问题,提出了一种智能混合建模方法,在构建基于区域法的管外传热计算模型的基础上,利用该模型产生的数据,设计构造了针对管外传热计算的神经网络模型。利用该模型与管内反应过程模型相耦合,实现对乙烯裂解炉辐射段的智能混合建模与模拟。结合工业实际算例,验证了基于机器学习和机理模型的智能混合建模的可行性,裂解产物预测精度良好,且混合模型可以大大缩短计算时间,更加符合工业计算的要求。     

Practical improvements to autocovariance least‐squares

Identifying disturbance covariances from data is a critical step in estimator design and controller performance monitoring. Here, the autocovariance least‐squares (ALS) method for this identification is examined. For large industrial models with poorly observable states, the process noise covariance is high dimensional and the optimization problem is poorly conditioned. Also, weighting the least‐squares problem with the identity matrix does not provide minimum variance estimates. Here, ALS method to resolve these two challenges is modified. Poorly observable states using the singular value decomposition (SVD) of the observability matrix is identified and removed, thus decreasing the computational time. Using a new feasible‐generalized least‐squares estimator that approximates the optimal weighting from data, the variance of the estimates is significantly reduced. The new approach on industrial data sets provided by Praxair is successfully demonstrated. The disturbance model identified by the ALS method produces an estimator that performs optimally over a year‐long period. © 2015 American Institute of Chemical Engineers AIChE J, 61: 1840–1855, 2015     

pH‐thermoreversible hydrogels. I. Synthesis and swelling behaviors of the (N‐isopropylacrylamide‐co‐acrylamide‐co‐2‐hydroxyethyl methacrylate) copolymeric hydrogels

A series of pH‐thermoreversible hydrogels that exhibited volume phase transition was synthesized by various molar ratios of N‐isopropylacrylamide (NIPAAm), acrylamide (AAm), and 2‐hydroxyethyl methacrylate (HEMA). The influence of environmental conditions such as temperature and pH value on the swelling behavior of these copolymeric gels was investigated. Results showed that the hydrogels exhibited different equilibrium swelling ratios in different pH solutions. Amide groups could be hydrolyzed to form negatively charged carboxylate ion groups in their hydrophilic polymeric network in response to an external pH variation. The pH sensitivities of these gels also depended on the AAm content in the copolymeric gels; thus the greater the AAm content, the higher the pH sensitivity. These hydrogels, based on a temperature‐sensitive hydrogel, demonstrated a significant change of equilibrium swelling in aqueous media between a highly solvated, swollen gel state and a dehydrated network response to small variations of temperature. pH‐thermoreversible hydrogels were used for a study of the release of a model drug, caffeine, with changes in temperature. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 71: 221–231, 1999     

Neural and Hybrid Neural Modeling and Control of Fed‐Batch Fermentation for Streptokinase: Comparative Evaluation under Nonideal Conditions

Fermentations involving competition between two or more kinds of cells under nonideal conditions show complex profiles that are sensitive to the extra‐cellular environment. These fermentations therefore require accurate and rapid on‐line data acquisition and control. However, both on‐line measurements and modelling are difficult and expensive for large bioreactors, thus limiting the usefulness of model‐based control. While neural networks offer an alternative, they require extensive training and can be difficult to optimize for large arrays. Hybrid networks combining a few neural networks with some mathematical equations offer a good compromise. The possibility of using a hybrid model for simulation‐cum‐control has been examined here for the fed‐batch production of streptokinase. Under noideal conditions, hybrid neural models outperformed both mathematical models and arrays of neural networks, thus suggesting their viability for large‐scale fermentation monitoring and control.     

Colour‐appearance modeling using feedforward networks with Bayesian regularization method— Part I: Forward model

In this article, a method of predicting colour appearance (from colorimetric attributes to colour‐appearance attributes, i.e., forward model) using an artificial neural network is presented. The neural network model developed is a multilayer feedforward neural network model for predicting colour appearance (FNNCAM for short). The model was trained by LUTCHI colour‐appearance datasets. The Levenberg–Marquardt algorithm is incorporated into the back‐propagation procedure to accelerate the training of FNNCAM and the Bayesian regularization method is applied to the training of neural networks to improve generalization. The results of FNNCAM obtained are quite promising. © 2000 John Wiley & Sons, Inc. Col Res Appl, 25, 424–434, 2000     

Dynamic On‐Line Reoptimization Control of a Batch MMA Polymerization Reactor Using Hybrid Neural Network Models

A hybrid neural network model based on‐line reoptimization control strategy is developed for a batch polymerization reactor. To address the difficulties in batch polymerization reactor modeling, the hybrid neural network model contains a simplified mechanistic model covering material balance assuming perfect temperature control, and recurrent neural networks modeling the residuals of the simplified mechanistic model due to imperfect temperature control. This hybrid neural network model is used to calculate the optimal control policy. A difficulty in the optimal control of batch polymerization reactors is that the optimization effort can be seriously hampered by unknown disturbances such as reactive impurities and reactor fouling. With the presence of an unknown amount of reactive impurities, the off‐line calculated optimal control profile will be no longer optimal. To address this issue, a strategy combining on‐line reactive impurity estimation and on‐line reoptimization is proposed in this paper. The amount of reactive impurities is estimated on‐line during the early stage of a batch by using a neural network based inverse model. Based on the estimated amount of reactive impurities, on‐line reoptimization is then applied to calculate the optimal reactor temperature profile for the remaining time period of the batch reactor operation. This approach is illustrated on the optimization control of a simulated batch methyl methacrylate polymerization process.     

Melt index prediction by RBF neural network with an ICO‐VSA hybrid optimization algorithm

Melt index (MI) is considered as one of the most significant parameter to determine the quality and the grade of the practical polypropylene polymerization products. A novel ICO‐VSA‐RNN (RBF neural network with ICO‐VSA algorithm) MI prediction model is proposed based on radial basis function (RBF) neural network and improved chaos optimization (ICO), and variable‐scale analysis (VSA), where the ICO is first added and then combined with the VSA to overcome the defects of ICO and VSA, then the parameters of the RBF neural network are optimized with them. At last, the RBF neural network model for MI prediction model is developed. Further researches on the optimal RBF neural network model of MI prediction are carried out with the data from a real industrial plant, and the prediction results show that the performance of this prediction model is much better than the RBF neural network model without optimization. © 2012 Wiley Periodicals, Inc. J Appl Polym Sci, 2012     

Integrated simulations of structural performance,molding process,and warpage for gas‐assisted injection‐molded parts. III. Simulation of cyclic,transient variations in mold wall temperatures

Whether it is feasible to perform an integrated simulation for structural analysis, process simulation, as well as warpage calculation based on a unified CAE model for gas‐assisted injection molding (GAIM) is a great concern. In the present study, numerical algorithms based on the same CAE model used for process simulation regarding filling and packing stages were developed to simulate the cooling phase of GAIM considering the influence of the cooling system. The cycle‐averaged mold cavity surface temperature distribution within a steady cycle is first calculated based on a steady‐state approach to count for overall heat balance using three‐dimensional modified boundary element technique. The part temperature distribution and profiles, as well as the associated transient heat flux on plastic–mold interface, are then computed by a finite difference method in a decoupled manner. Finally, the difference between cycle‐averaged heat flux and transient heat flux is analyzed to obtain the cyclic, transient mold cavity surface temperatures. The analysis results for GAIM plates with semicircular gas channel design are illustrated and discussed. It was found that the difference in cycle‐averaged mold wall temperatures may be as high as 10°C and within a steady cycle, part temperatures may also vary ∼ 15°C. The conversion of gas channel into equivalent circular pipe and further simplified to two‐node elements using a line source approach not only affects the mold wall temperature calculation very slightly, but also reduces the computer time by 95%. This investigation indicates that it is feasible to achieve an integrated process simulation for GAIM under one CAE model, resulting in great computational efficiency for industrial application. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 71: 339–351, 1999     

Drying Kinetics Prediction of Solid Waste Using Semi‐Empirical and Artificial Neural Network Models

The drying process of organic solid waste is investigated, based on an experimental study involving its drying kinetics. The experiments were conducted in a thin‐layer fixed‐bed dryer under various operational conditions. The problem of selecting the best fit for solid waste moisture content as a function of time is addressed as well, using artificial neural network (ANN) models and four well‐known drying kinetics correlations commonly applied to biological materials. According to the statistical analysis employed, the simulations showed good results for the ANN, and the Overhults model provided optimum agreement with experimental data among all other models evaluated. Empirical correlations between the Overhults model parameters and the drying operational conditions using nonlinear regression techniques were determined.     

Using hybrid neural networks in scaling up an FCC model from a pilot plant to an industrial unit

The scaling up of a pilot plant fluid catalytic cracking (FCC) model to an industrial unit with use of artificial neural networks is presented in this paper. FCC is one of the most important oil refinery processes. Due to its complexity the modeling of the FCC poses great challenge. The pilot plant model is capable of predicting the weight percent of conversion and coke yield of an FCC unit. This work is focused in determining the optimum hybrid approach, in order to improve the accuracy of the pilot plant model. Industrial data from a Greek petroleum refinery were used to develop and validate the models. The hybrid models developed are compared with the pilot plant model and a pure neural network model. The results show that the hybrid approach is able to increase the accuracy of prediction especially with data that is out of the model range. Furthermore, the hybrid models are easier to interpret and analyze.     

Gas Analysis by In Situ Combustion in Heavy‐Oil Recovery Process: Experimental and Modeling Studies

Enormous efforts have been made to facilitate produced‐gas analyses by in situ combustion implication in heavy‐oil recovery processes. Robust intelligence‐based approaches such as artificial neural network (ANN) and hybrid methods were accomplished to monitor CO₂/O₂/CO. Implemented optimization approaches like particle swarm optimization (PSO) and hybrid approach focused on pinpointing accurate interconnection weights through the proposed ANN model. Solutions acquired from the developed approaches were compared with the pertinent experimental in situ combustion data samples. Implication of hybrid genetic algorithm and PSO in gas analysis estimation can lead to more reliable in situ combustion quality predictions, simulation design, and further plans of heavy‐oil recovery methods.     

3‐D full‐loop simulation of an industrial‐scale circulating fluidized‐bed boiler

Three‐dimensional (3‐D) simulations using an Eulerian multiphase model were employed to explore flow behaviors in a full‐loop industrial‐scale CFB boiler with and without fluidized‐bed heat exchanger (FBHE), where three solids phases were employed to roughly represent the polydisperse behavior of particles. First, a simulation of the boiler without FBHE is implemented to evaluate drag models, in terms of pressure profiles, mixing behaviors, radial velocity profiles, etc. Compared to the conventional model, the simulation using the energy‐minimization multiscale (EMMS) model successfully predicts the pressure profile of the furnace. Then, such method is used to simulate the boiler with FBHE. The simulation shows that solid inventory in the furnace is underpredicted and reduced with an increase of the valve opening, probably due to the underevaluated drag for FBHE flows. It is suggested to improve EMMS model which is now based on a single set of operating parameters to match with the full‐loop system. © 2012 American Institute of Chemical Engineers AIChE J, 59: 1108–1117, 2013     

Colour‐appearance modeling using feedforward networks with Bayesian regularization method. Part II: Reverse model

In Part I of this article, the development of a multilayer perceptrons feedforward artificial neural network model to predict colour appearance from colorimetric values was reported. Bayesian regularization was employed for the training of the network. In this part of the article, the reverse model, that is, the perdition of colorimetric values from the colour appearance attributes is reported using the same neural network design methodology developed in Part I. This study should contribute to the building of an artificial neural network–based colour appearance prediction, both forward and reverse, using the most comprehensive LUTCHI colour appearance data sets for training and testing. Good prediction accuracy and generalization ability were obtained using the neural networks built in the study. Because the neural network approach is of a black‐box type, colour appearance prediction using this method should be easier to apply in practice. © 2002 Wiley Periodicals, Inc. Col Res Appl, 27, 116–121, 2002; DOI 10.1002/col.10030