Desiccant wheel system modeling improvement using multiple population genetic algorithm training of neural network

Since the physical modeling method for the desiccant wheel system (DWS) is complex and costly for calculations, the modeling method based on neural network (NN) gains attention for its simplicity and effectiveness. The previous NN models of DWS are mostly based on backpropagation (BP) NN and adopt the gradient searching method to obtain the weights and thresholds. However, the gradient searching method results in “overfitting” easily. In this paper, a novel neural network training algorithm, trainmpga, is proposed. The algorithm searches the optimal weights and thresholds of NN by making use of the multiple population genetic algorithm, thereby conquering the “overfitting” of the gradient searching method and the “prematurity” of the genetic algorithm. Meanwhile, related configurations of NN, such as parameters and framework, are studied. Finally, the proposed modeling method trainmpga proves to have high training and prediction accuracy in comparison to the training algorithms in the MatLab toolbox and has good application prospects.     

Miscibility,morphology, structure,and properties of porous cellulose–soy protein isolate hybrid hydrogels

Porous hybrid hydrogels were fabricated by mixing cellulose (CEL) and soy protein isolate (SPI) solutions, followed by crosslinking with epichlorohydrin. Their miscibility, morphology, structure, and properties were investigated by wide‐angle X‐ray diffraction, thermogravimetric analysis, scanning electron microscopy, dynamic mechanical analysis, rheological measurement, and swelling tests. The results show that CEL performed as a “scaffold” of pore walls and contributed to the good mechanical properties, while SPI performed the role of an “extender” of pore size and was responsible for the high water absorbency. The incorporation of CEL (stiff chains) and SPI (hydrophilic groups) in the hybrid hydrogel constructed the porous structure. This work provides a method for the fabrication of hydrogels with porous structure through the combination of a stiff material as a “scaffold” and a hydrophilic material as an “extender.” © 2016 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2016 , 133, 43853.     

Modelling of A Fluidized Bed Drier Using Artificial Neural Network

ABSTRACT

Proper modelling of a fluidized bed drier (FBD) is important to design model based control strategies. A FBD is a non-linear multivariable system with non-minimum phase characteristics. Due to the complexities in FBD conventional modelling techniques are cumbersome. Artificial neural network (ANN) with its inherent ability to “learn” and “absorb” non-linearities, presents itself as a convenient tool for modelling such systems.

In this work, an ANN model for continuous drying FBD is presented. A three layer fully connected feedfordward network with three inputs and two outputs is used. Backpropagation learning algorithm is employed to train the network. The training data is obtained from computer simulation of a FBD model from published literature. The trained network is evaluated using randomly generated data as input and observed to predict the behaviour of FBD adequately.     

青霉素发酵过程中的混合建模

由于微生物发酵过程机理的复杂性和高度非线性,建立发酵过程的精确模型具有一定难度。传统的动力学模型预测虽然会与实际输出有一定偏差,但它在某种程度上反映了过程机理;神经网络建模方法属于"黑箱"方法,建模过程中没有用到任何先验知识,有一定的预测效果;因此单一的建模方法往往会不具备其他建模方法的优势。本文以传统的发酵动力学模型为基础,结合RBF神经网络进行混合建模的"灰箱"建模方法是一种比较好的建模思路,可以取得较满意的软测量效果。     

Multidimensional non-orthogonal wavelet-sigmoid basis function neural network for dynamic process fault diagnosis

Dealing with multidimensional problems has been the “bottle-neck” for implementing wavenets to process systems engineering. To tackle this problem, a novel multidimensional wavelet (MW) is presented with its rigorously proven approximation theorems. Taking the new wavelet function as the activation function in its hidden units, a new type of wavenet called multidimensional non-orthogonal non-product wavelet-sigmoid basis function neural network (WSBFN) model is proposed for dynamic fault diagnosis. Based on the heuristic learning rules presented by authors, a new set of heuristic learning rules is presented for determining the topology of WSBFNs. The application of the proposed WSBFN is illustrated in detail with a dynamic hydrocracking process.     

Numerical solution of equations arising in energy transport in non-equilibrium chemically reacting systems

A two-point boundary value method is presented for avoiding the numerical instability which can occur when initial value methods are used for solving the non-linear differential equations describing transport of heat and mass in chemically reacting systems near equilibrium. The instability arises primarily owing to the slight computational errors that arise when the forward and reverse reaction-rate terms are large compared to the net rate. The method used to avoid this instability is to convert the continuity of species equation containing the reaction-rate terms into a second order differential equation in such a manner that small errors in computing the reaction rates do not appreciably affect the solution. Though equations leading to this instability have sometimes been referred to in the past as “stiff,” today this term most commonly refers to quite a different situation, namely, a system of equations where at any point the ratio of the largest to smallest eigenvalue is very large. Thus, in this work the term “stiff” will be replaced by “instability.” As an example of how to circumvent this instability, the method is applied to the problem of heat transfer through a reacting gaseous mixture confined in a parallel plate conductivity cell. It is demonstrated that this case presents a more stringent test of the applicability of the method than the case of convective heat transfer.     

Hybrid modelling and self-learning system for dextrose crystallization process

This paper deals with the modelling of a continuous cooling column crystallizer. An accurate model of this system is needed for complex process control. The investigated system consists on dextrose monohydrate in aqueous solution. An adaptive hybrid model is presented. The model consists of two parts: the phenomenological model, expressed by a set of differential algebraic equations, and a neural network, based on historical data, developed by the fuzzy ARMAP technique. The empirical part of the hybrid model is aimed at eliminating the deviations of the prediction of the phenomenological model caused mainly by incrustations over the surface of the cooling coils located along the column crystallizer. The model is adaptive since neural network parameters are updated by a self-learning system (SLS) based on the acquired process data storage of the DCS of an industrial plant. Firstly, the hybrid model was implemented by using the data of a three-month campaign of the crystallizer, then the self-learning technique was checked on site in a subsequent campaign.     

Flexible on‐line Modeling and Control of pH in Waste Neutralization Reactors

The control of pH in waste neutralization processes presents a challenging highly nonlinear and time‐varying problem in which the reactor also suffers from inaccessible state information. The ability to characterize the changing dynamics of such reactors is essential to the success of advanced control schemes for these applications. In this work, flexible on‐line modeling of a pH reactor simulating nonstationary behavior was studied. This entailed a comparison of the most popular connectionist learning algorithm, the “Widrow‐Hoff delta rule”, with a classical tool in adaptive identification and control, recursive least squares (RLS). The modeling was pursued within the framework of neural networks using the ADALINE neural network. Further, two heuristically defined first‐principles‐based transforms were investigated for providing “general globally linearizing” information to the ADALINE. The comparisons of the learning algorithms for different neural network information vectors has led to a critical understanding of the flexibility of each algorithm for on‐line learning of the diverse process gain characteristics encountered in pH reactors.     

利用结构逼近式混合神经网络实现间歇反应器的建模

提出了一种新的混合神经网络建模方法——结构逼近式混合神经网络。基于此结构建立的混合神经网络可以充分利用已知非线性系统的结构信息,使神经网络“灰盒”化,更好地解释和描述系统各变量间的因果关系,从而提高网络的建模精度和模型的可靠性。本文介绍了这类神经网络的基本特性、拓扑结构和训练方法。报告了一个典型放热液相二级平行间歇反应的建模过程;并针对间歇反应过程测量滞后的情况,与两种不同的混合神经网络模型作了比较,仿真和比较结果证明了方法的有效性。     

Energy absorption characteristics of polymeric foams used as cushioning materials

The “Efficiency of Energy Absorption” (or “Efficiency”) and “Ideality” parameters were evaluated for several plastic foams and were found to be very useful in choosing an appropriate cushion for the protection of a specific product. The maxima in these parameters were found to be in the same range of stresses, when derived from dynamic measurements or from predicted stress-strain curves based on previously proposed “Reference” and “Modified Boltzman Superposition” Models. For the rate independent foams the maxima in these parameters do not change with strain rate and can therefore be derived from slow, constant rate (“static”) experiments. For rate dependent foams however, the maxima from “static” measurements were found to be in a lower range of stresses than those derived from dynamic ones. As a result, slow rate compressive measurements do not predict well the behavior of the foams during impact and the use of the “Reference” and “Modified Boltzman Superposition” Models is required for good predictions. It was found that the suffer PS and PE foams attain maximum Efficiency and Ideality at higher stresses than the more flexible PUR foams.     

A Mathematical Model of SOFC: A Proposal

The mathematical model of the solid oxide fuel cell (SOFC) is presented. The new approach for modeling the voltage of SOFC is proposed. Electrochemical, thermal, electrical, and flow parameters are collected in the 0D mathematical model. The aim was to combine all cell working conditions in as a low number of factors as possible and to have the factors relatively easy to determine. A validation process for various experimental data was made and adequate results are shown. The presented model was validated for various fuel mixtures in relatively wide ranges of parameters as well as for various cell design parameters (e.g. electrolyte thickness, anode porosity, etc.). A distinction is made between the “design‐point” and “off‐design operation”.     

Introducing two simple models for predicting fiber‐reinforced asphalt concrete behavior during longitudinal loads

Scientists and engineers are constantly trying to improve the performance of asphalt pavements. Modification of the asphalt binder is one approach taken to improve pavement performance. The idea of using fibers to improve the behavior of materials is an old suggestion, so different researchers reported the results of adding a large variety of fibers to asphalt concrete (AC) as fiber‐reinforced asphalt concrete (FRAC). However, there are few comments about the mechanism of reinforcement and fiber performance in the inner structure of AC and/or exposing some models to predict fiber recital as a modifier in FRAC. So this article is going to introduce two simple models for predicting FRAC behavior during longitudinal loads. The former is called “Slippage Theory” and the latter is “Equal Cross‐Section.” Finally, four types of fibers (glass, nylon 6.6, polypropylene, and polyester) were used in AC to evaluate the two theories. “Marshall Test,” as stability and flow outcomes, and “Specific Gravity” were carried out on specimens in the next stages followed by an artificial neural network (ANN), which was developed in the system to recognize important fiber parameters effective in the FRAC specifications. In the end, the two theories predicted each fiber performance in FRAC as well as ANN. © 2008 Wiley Periodicals, Inc. J Appl Polym Sci, 2008     

A Dynamic Simulation Tool for the Battery‐Hybrid Hydrogen Fuel Cell Vehicle

This paper describes a dynamic fuel cell vehicle simulation tool for the battery‐hybrid direct‐hydrogen fuel cell vehicle. The emphasis is on simulation of the hybridized hydrogen fuel cell system within an existing fuel cell vehicle simulation tool. The discussion is focused on the simulation of the sub‐systems that are unique to the hybridized direct‐hydrogen vehicle, and builds on a previous paper that described a simulation tool for the load‐following direct‐hydrogen vehicle. The configuration of the general fuel cell vehicle simulation tool has been previously presented in detail, and is only briefly reviewed in the introduction to this paper. Strictly speaking, the results provided in this paper only serve as an example that is valid for the specific fuel cell vehicle design configuration analyzed. Different design choices may lead to different results, depending strongly on the parameters used and choices taken during the detailed design process required for this highly non‐linear and n‐dimensional system. The primary purpose of this paper is not to provide a dynamic simulation tool that is the “final word” for the “optimal” hybrid fuel cell vehicle design. The primary purpose is to provide an explanation of a simulation method for analyzing the energetic aspects of a hybrid fuel cell vehicle.     

基于数据驱动的卷积神经网络电容层析成像图像重建

针对电容层析成像技术的图像重建问题，提出了基于数据驱动的卷积神经网络图像重建方法。根据气固两相流的流型特点，通过数值模拟的方法随机生成了60000组介质分布图像，并利用有限元法计算了与之对应的电容向量，从而建立了一个“电容向量-介质分布”数据集；然后根据电容层析成像图像重建特点建立了卷积神经网络模型，对数据集中的训练集进行学习和训练，并利用测试集对训练结果进行了验证与评价。在此基础上，对获得的ECT图像重建卷积神经网络模型进行了静态实验和流化床测试实验研究。模拟和实验结果表明：所建立的卷积神经网络能较好地实现ECT图像重建，可直接用于流化床内的颗粒浓度分布测量。     

Mechanical properties of glass fiber/organic fiber mixed–mat reinforced thermoplastic composites

The purpose of this study is to investigate the influence of different types of fibers on the mechanical properties of hybrid composite materials. Long and short glass fibers (GF) and different types of organic fibers, viz. aramid fiber, DuPont Kevlar‐49 (KF), liquid crystalline polymer (LCP), and vinylon (VF) in hybrid composites, were used to reinforced the high density polyethylene (HDPE) matrix. The long fiber hybrid composites were prepared in a “fiber separating and flying machine,” while the short fiber hybrid composites were prepared in an “elastic extruder.” The total amount of fibers used in both long and short fiber hybrid composites was fixed at 20 vol%. The influence of fiber content, length, and mixing ratio on mechanical properties, such as tensile, bending, Izod and high rate impact strength, as well as viscoelastic propertics in the solid state, was studied. Fracture surfaces of the materials were also examined using a scanning electron microscopy.     

HYBRID NEURAL APPROACHES FOR MODELLING DRYING PROCESSES FOR PARTICULATE SOLIDS

ABSTRACT

This work presents methods for synthesizing drying process models for particulate solids that combine prior knowledge with artificial neural networks. The inclusion of prior knowledge is investigated by developing two applications with the data from two indirect rotary steam dryers. The first application consisted in the modelling of the drying process of soya meal in a batch indirect rotary dryer, The external and internal mass transfer resistances were associated in the hidden layer of the network to linear and sigmoidal nodes, respectively. The second application consisted in the modelling of the drying process of soya meal in a continuos indirect rotary dryer. The model was constructed using the Semi-parametric Design Approach. The model predicts the evolution of solid moisture content and temperature as a function of the solid position in the dryer. The results show that the hybrid model performs better than the pure “ black box” neural network and default models. They also shows that prior knowledge enhances the extrapolation capabilities of a neural network model,     

Modelling and control of different types of polymerization processes using neural networks technique: A review

Polymerization process can be classified as a nonlinear type process since it exhibits a dynamic behaviour throughout the process. Therefore, it is highly complicated to obtain an accurate mechanistic model from the nonlinear process. This predicament always been a “wall” to researchers to be able to devise an optimal process model and control scheme for such a system. Neural networks have succeeded the other modelling and control methods especially in coping with nonlinear process due to their very conciliate characteristics. These characteristics are further explained in this work. The predicament that is encountered by researchers nowadays is lack of data which consequently lead to an imprecise mechanistic model that scarcely conforms to the desired process. The implementations of the neural network model not only restrict to polymerization reactor but to other difficult‐to‐measure parameters such as polymer quality, polymer melts index and mixture of initiators. This work is aimed to manifest ascendancy of neural networks in modelling and control of polymerization process.     

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     

Synthesis of optimal energy recovery networks using discrete methods

A flexible modular program system for the synthesis of optimal energy recovery networks is described. Process networks are generated to meet both stream temperature and pressure specifications. The system makes use of a price based decomposition algorithm for sub-process integration. It incorporates heuristic decision making and employs a “branch and bound” combinatorial technique for optimization of network configuration. The ability of the system to generate very realistic process designs is demonstrated by its application to the design of an energy recovery system for an ethylene plant with its associated refrigeration facility.