Model Reduction in Emulsion Polymerization Using Hybrid First Principles/Artificial Neural Networks Models, 2

Summary: A “series” hybrid model based on material balances and artificial neural networks to predict the evolution of weight average molecular weight, , in semicontinuous emulsion polymerization with long chain branching kinetics is presented. The core of the model is composed by two artificial neural networks (ANNs) that calculate polymerization rate, R_p, and instantaneous weight‐average molecular weight, from reactor process variables. The subsequent integration of the material balances allowed to obtain the time evolution of conversion and , along the polymerization process. The accuracy of the proposed model under a wide range of conditions was assessed. The low computer‐time load makes the hybrid model suitable for optimization strategies.

Effect of the monomer feed rate on .     

Hybrid neural modeling of bioprocesses using functional link networks

The objective of this work was to develop a model for an extractive ethanol fermentation in a simple and rapid way. This model must be sufficiently reliable to be used for posterior optimization and control studies. A hybrid neural model was developed, combining mass and energy balances with neural networks, which describe the process kinetics. To determine the best model, two structures of neural networks were compared: the functional link networks and the feedforward neural networks. The two structures are shown to describe well the process kinetics, and the advantages of using the functional link networks are discussed.     

Adaptive optimal control of fed-batch alcoholic fermentation

An adaptive control scheme is developed for the optimization of a fed-batch ethanol production process. The fermentation process is modeled by an hybrid neural model combining mass balance equations and neural networks, used to represent the kinetic rates. The networks used, the functional link networks (FLN), allow the linear estimation of their parameters; this enables the re-estimation of the parameters at each sampling time, and thus the development of an adaptive optimal control scheme.     

Phenomenological and neural-network modeling of cephalosporin C production bioprocess

Cephalosporin C production process withCephalosporium acremonium ATCC 48272 in synthetic medium was investigated and the experimental results allowed the development of a mathematical model describing the process behavior. The model was able to explain fairly well the diauxic phenomenon, higher growth rate during the glucose-consumption phase, and the production occurring only in the sucrose-consumption phase. Moreover, the process was simulated utilizing the neural-networks technique. Two feed-forward neural-networks with one hidden layer were employed. Both models, phenomenological and neural-networks based, satisfactorily describe the bioprocess. The difficulties in determining kinetic parameters are avoided when neural networks are utilized.     

Hybrid neural network model of an industrial ethanol fermentation process considering the effect of temperature

In this work a procedure for the development of a robust mathematical model for an industrial alcoholic fermentation process was evaluated. The proposed model is a hybrid neural model, which combines mass and energy balance equations with functional link networks to describe the kinetics. These networks have been shown to have a good nonlinear approximation capability, although the estimation of its weights is linear. The proposed model considers the effect of temperature on the kinetics and has the neural network weights reestimated always so that a change in operational conditions occurs. This allow to follow the system behavior when changes in operating conditions occur.     

Nanopore generation in hybrid polycyanurate/poly(ε-caprolactone) thermostable networks

Novel nanoporous thermosetting films were obtained from thermostable polycyanurate (PCN)-based hybrid networks synthesized by polycyclotrimerization of cyanate ester of bisphenol E in the presence of a modifier reactive toward cyanate groups, i.e. dihydroxy-telechelic poly(ε-caprolactone) (PCL). The nanoporous structure was generated in PCN/PCL hybrid networks after extraction of unreacted free PCL sub-chains which were not chemically incorporated into the PCN cross-linked framework. Structure–property relationships for precursory and porous PCN/PCL hybrid networks were investigated using a large array of physico-chemical techniques. The porosity associated with the networks after extraction was more particularly evaluated by SEM and DSC-based thermoporometry: pore sizes around 10–90 nm were determined along with pore volumes as high as about 0.3 cm³ g⁻¹. Density and dielectric measurements strongly suggested the occurrence of closed pore structures. Due to their high thermal stability as investigated by TGA, nanoporous PCN/PCL hybrid cross-linked films could be considered as promising materials for potential applications as thermostable membranes.     

A hybrid neuroal network algorithm for on-line state inference that accounts for differences in inoculum of Cephalosporium acremonium in fed-batch fermentors

One serious difficulty in modeling a fermentative process is the forecasting of the duration of the lag phase. The usual approach to model biochemical reactors relies on first-principles, unstructured mathematical models. These models are not able to take into account changes in the process response caused by different incubation times or by repeated fed batches. Toover come this problem, we have proposed a hybrid neural network algorithm. Feedforward neural networks were used to estimate rates of cell growth, substrate consumption, and product formation from on-line measurements during cephalosporin C production. These rates were included in the mass balance equations to estimate key process variables: concentrations of cells, substrate, and product. Data from fed-batch fermentation runs in a stirred aerated bioreactor employing the microorganism Cephalosporium acremonium ATCC 48272 were used. On-line measurements strongly related to the mass and activity of the cells used. They include carbon dioxide and oxygen concentrations in the exhausted gas. Good results were obtained using this approach.     

基于神经网络的多元稀土萃取组分含量的软测量

提出了一种应用RBF神经网络建立多组分稀土萃取平衡模型的方法,通过与萃取过程物料平衡模型相结合给出了在线预测稀土申级萃取分离生产过程中各组分含量的软测量方法。通过现场操作运行实测数据的建模仿真验证,表明这种混合软测量方法是有效的。     

Model Reduction in Emulsion Polymerization Using Hybrid First‐Principles/Artificial Neural Network Models

A first‐principles mathematical model for emulsion polymerization was reduced by using a hybrid mathematical model composed by artificial neural networks (ANN) and material balances. The goal was to have an accurate model that may be integrated fast enough to be used for online optimization purposes. In the reduced model the polymerization rate and the instantaneous weight‐average molecular weight were calculated by means of artificial neural networks. These ANNs were incorporated to first‐principles material balances. The accuracy of the reduced model under a wide range of conditions was assessed. Savings in computer time were achieved by using the reduced model, which makes it suitable for online optimization purposes.

Effect of the temperature on the cumulative weight‐average molecular weight: first principles mathematical model (—); (ANN2) and hybrid model predictions: (▵) 50 °C, (▪) 60 °C^(training), (▿) 70 °C^(validation), (•) 80 °C, (○) 90 °C.     

Analyses of the temperature and pH effects on the complexation of magnesium and calcium in human blood plasma: an approach using artificial neural networks

The temperature and pH effects on the equilibrium of a blood plasma model have been studied on the basis of artificial neural networks. The proposed blood plasma was modeled considering two important metals, calcium and magnesium, and six ligands, namely, alanate, carbonate, citrate, glycinate, histidinate and succinate. A large data set has been used to simulate different concentrations of magnesium and calcium as a function of temperature and pH and these data were used for training the neural network. The proposed model allowed different types of analyses, such as the effects of pH on calcium and magnesium concentrations, the competition between calcium and magnesium for ligands and the effects of temperature on calcium and magnesium concentrations. The model developed was also used to predict how the variation of calcium concentration can affect magnesium concentrations. A comparison of neural network predictions against experimental data produced errors of about 3%. Moreover, in agreement with experimental measurements (Wang et al. in Arch. Pathol. 126:947–950, 2002; Heining et al. in Scand. J. Clin. Lab. Invest. 43:709–714, 1983), the artificial neural network predicted that calcium and magnesium concentrations decrease when pH increases. Similarly, the magnesium concentrations are less sensitive than calcium concentrations to pH changes. It is also found that both calcium and magnesium concentrations decrease when the temperature increases. Finally, the theoretical model also predicted that an increase of calcium concentrations will lead to an increase of magnesium concentration almost at the same rate. These results suggest that artificial neural networks can be efficiently applied as a complementary tool for studying metal ion complexation, with especial attention to the blood plasma analysis. Figure Artificial neural networks for predicting the behavior of calcium and magnesium as a function of pH and temperature in human blood plasma     

Neural networks applied in chemistry. I. Determination of the optimal topology of multilayer perceptron neural networks

Artificial neural networks (ANNs) are comparatively straightforward to understand and use in the analysis of scientific data. However, this relative transparency may encourage their use in an uncritical, and therefore possibly unproductive, fashion. The geometry of a network is among the most crucial factors in the successful deployment of network tools; in this review, we cover methods that can be used to determine optimum or near‐optimum geometries. These methods of determining neural network architecture include the following: (i) trial and error, in which architectures chosen semirandomly are tested and modified by the user; (ii) empirical or statistical methods, in which an ANN's internal parameters are adjusted based on the model's performance; (iii) hybrid methods, such as fuzzy inference; (iv) constructive and/or pruning algorithms, that add and/or remove neurons or weights from an initial architecture, respectively, based on a predefined link between architecture and ANN performance; (v) evolutionary strategies, which search the topology space using genetic operators to vary the neural network parameters. Several case studies illustrate the development of neural network models for applications in chemistry and chemical engineering. Copyright © 2011 John Wiley & Sons, Ltd.     

Software sensors for monitoring of a solid waste composting process

Process identification for composting of tobacco solid waste in an aerobic, adiabatic batch reactor was carried out using neural network-based models which utilized the nonlinear finite impulse response and nonlinear autoregressive model with exogenous inputs identification methods. Two soft sensors were developed for the estimation of conversion. The neural networks were trained by the adaptive gradient method using cascade learning. The developed models showed that the neural networks could be applied as intelligent software sensors giving a possibility of continuous process monitoring. The models have a potential to be used for inferential control of composting process in batch reactors. Presented at the 33rd International Conference of the Slovak Society of Chemical Engineering, Tatranské Matliare, 22–26 May 2006.     

Preparation and trans–cis isomerization of azobenzene-containing TiO2/ormosils hybrid films derived at a low temperature sol–gel process for photonic applications

Azobenzene-containing TiO₂/γ-glycidoxypropyltrimethoxysilane and methyltrimethoxysilane hybrid films are prepared by combining a low temperature sol–gel process and a spin-coating technique. The trans–cis isomerization of azobenzene small molecules inside organic–inorganic hybrid films is induced by a photoirradiation under UV light. It is noted that below a baking temperature of 150 °C is necessary for the hybrid films in optical storage or optical switching applications. The change of the refractive index and thickness of the hybrid films with real-time heating temperature are observed by using a prism coupling technique. The structural properties of the hybrid films are also characterized and investigated by thermal gravimetric analysis, Fourier transform infrared spectroscopy and X-ray photoelectron spectroscopy. The results indicate that the as-prepared hybrid films might allow directly integrating the optical storage or optical switching devices with the waveguide devices on the same chip.     

Study of the thermal degradation of bioactive sol–gel coatings for the optimization of its curing process

A set of materials has been prepared by sol–gel process containing different quantities of hydroxyapatite (0, 2.5 and 5% HAp w/w) using as silica precursors glycidyloxypropyltrimethoxysilane (GPTMS) and triethoxyvinylsilane (VTES). In order to optimize the curing process to obtain sintherized systems (inorganic network) or hybrid systems (organic–inorganic) a TG and FTIR studies have been developed and degradation kinetic triplet parameters were obtained (the activation energy, pre-exponential factor, and function of degree of conversion). The kinetic study was analyzed by means of an integral isoconversional non-isothermal procedure (model free), and the kinetic model was determined by the Coats–Redfern method and through the compensation effect (IKR). All the systems followed the n = 6 kinetic model. The addition of HAp increases the thermal stability of the systems. The isothermal degradation was simulated from non-isothermal data, and the curing process could be defined to obtain the two types of materials. Temperature under 250 °C allows the formation of hybrids networks.     

人工神经网络在纸浆卡伯值光学定量分析中的应用

卡伯值 (硬度 )是纸浆的重要质量指标 ,是制浆过程控制的关键参数 .目前的测量方法包括化学分析法和光学分析法两大类型 ,国内大多数的制浆造纸厂采用离线的传统化学分析法来测定纸浆的卡伯值 ,需要比较长的时间 .而光学分析法因具有实时性好、精度和可靠性高等优点 ,已逐步用于卡伯值的在线测量和控制 .研究 [1] 发现 ,在 460～ 580 nm的可见光谱范围内 ,蒸煮液吸光度的变化可以表征纸浆中木素含量的变化 .本文将可见分光光谱技术应用于蒸煮液中木素含量的在线测量 ,根据蒸煮液在所选波段的吸光度来预测纸浆的卡伯值 ,建立纸浆卡伯值与蒸煮…     

Organic–inorganic hybrid membranes prepared from the sol–gel process of poly(butyleneadipate‐co‐terephthalate) and TiO2

Organic–inorganic hybrids based on poly(butyleneadipate‐co‐terephthalate)/titanium dioxide (PBAT/TiO₂) hybrid membranes were prepared via a sol–gel process. The PBAT/TiO₂ hybrid membranes were prepared for various PBAT/TiO₂ ratios. The resulting hybrids were characterized with a morphological structure, hydrophilicity, biodegradability, and thermal properties. The results showed that macrovoids underwent a transition into a sponge‐like membrane structure with the addition of TiO₂. After sol–gel transition, a strong interaction between the inorganic network and polymeric chains led to an increase in glass transition temperature (T_g), thermal degrading temperature, and hydrophilicity, and hence a higher biodegradability. According to X‐ray diffraction measurements of the crystal structure of the hybrid, the presence of TiO₂ did not change the crystal structure of PBAT. TiO₂ networks are uniformly dispersed into the PBAT matrix and no aggregation of TiO₂ networks in the hybrid membranes was observed through the small angle X‐ray scattering measurements. Thus, the sol–gel process of PBAT and TiO₂ can be used to prepare a hybrid with higher application temperature and faster biodegradation rate. Copyright © 2008 John Wiley & Sons, Ltd.     

Dynamic modelling of milk ultrafiltration by artificial neural network

Artificial neural networks (ANNs) have been used to dynamically model crossflow ultrafiltration of milk. It aims to predict permeate flux, total hydraulic resistance and the milk components rejection (protein, fat, lactose, ash and total solids) as a function of transmembrane pressure and processing time. Dynamic modelling of ultrafiltration performance of colloidal systems (such as milk) is very important for designing of a new process and better understanding of the present process. Such processes show complex non-linear behaviour due to unknown interactions between compounds of a colloidal system, thus the theoretical approaches were not being able to successfully model the process. In this work, emphasis has been focused on intelligent selection of training data, using few training data points and small network. Also it has been tried to test the ANN ability to predict new data that may not be originally available. Two neural network models were constructed to predict the flux/total resistance and rejection during ultrafiltration of milk. The results showed that there is an excellent agreement between the validation data (not used in training) and modelled data, with average errors less than 1%. Also the trained networks are able to accurately capture the non-linear dynamics of milk ultrafiltration even for a new condition that has not been used in the training process.