基于小波网络的非线性动态过程建模和在环氧氯丙烷生产过程中的应用

A type of wavelet neural network, in which the scale function is adopted only,is proposed in this paper for non-linear dynamic process modelling.Its network size is decreased significantly and the weight coefficients can be estimated by a linear algorithm.The wavelet neural network holds some advantages supeiior to other types of neural networks.First, its network structure is easy to specify based on its theoretical analysis and intuition.Secondly, network training does not rely on stochastic gradient type techniques and avoidd the problem of poor convergence or undesirable local minima.The excellent statistic properties of the weight parameter estimations can be proven here.Both theoretical analysis and simulation study show that the identification method is robust and reliable. Furthermore,a hybrid network structure incorporating first-principle knowledge and wavelet network is developed to solve a commonly existing problem in chemical production processes.Applications of the hybrid network to a practical production process demonstrates that model generalisation capability is significantly improved.     

输入训练神经网络的维数约简算法及其在化工过程建模中的应用

Many applications of principal component analysis （PCA） can be found in dimensionality reduction. But linear PCA method is not well suitable for nonlinear chemical processes. A new PCA method based on improved input training neural network （IT-NN） is proposed for the nonlinear system modelling in this paper. Momentum factor and adaptive learning rate are introduced into learning algorithm to improve the training speed of IT-NN. Contrasting to the auto-associative neural network （ANN）, IT-NN has less hidden layers and higher training speed. The effectiveness is illustrated through a comparison of IT-NN with linear PCA and ANN with experiments. Moreover, the IT-NN is combined with RBF neural network （RBF-NN） to model the yields of ethylene and propylene in the naphtha pyrolysis system. From the illustrative example and practical application, IT-NN combined with RBF-NN is an effective method of nonlinear chemical process modelling.     

通过数学规划方法设计具有混合结构的水回用网络

A new design method for a water-reusing network, with a hybrid structure, to reduce the complexity of the network and to minimize freshwater consumption, is proposed. The unique feature of the methodology proposed .in this article is to control the complexity of the water network by regulation of the control number in a water-reusing system. It combines the advantages of a conventional water-reusing network and a water-reusing net work with internal water mains. To illustrate the proposed method, a single contaminant system and a multiple contaminant system serve as examples of the problems.     

DOBD Algorithm for Training Neural Network: Part I. Method

Overfitting is one of the important problems that restrain the application of neural network. The traditional OBD (Optimal Brain Damage) algorithm can avoid overfitting effectively. But it needs to train the network repeatedly with low calculational efficiency. In this paper, the Marquardt algorithm is incorporated into the OBD algorithm and a new method for pruning network－the Dynamic Optimal Brain Damage (DOBD) is introduced. This algorithm simplifies a network and obtains good generalization through dynamically deleting weight parameters with low sensitivity that is defined as the change of error function value with respect to the change of weights. Also a simplified method is presented through which sensitivities can be calculated during training with a little computation. A rule to determine the lower limit of sensitivity for deleting the unnecessary weights and other control methods during pruning and training are introduced. The training course is analyzed theoretically and the reason why DOBD algorithm can obtain a much faster training speed than the OBD algorithm and avoid overfitting effectively is given.     

基于FCMISVM的自适应软测量建模算法及其在PX吸附分离过程中的应用

To overcome the problem that soft sensor models cannot be updated with the process changes, a soft sensor modeling algorithm based on hybrid fuzzy c-means (FCM) algorithm and incremental support vector machines (ISVM) is proposed. This hybrid algorithm FCMISVM includes three parts: samples clustering based on FCM algorithm, learning algorithm based on ISVM, and heuristic sample displacement method. In the training process, the training samples are first clustered by the FCM algorithm, and then by training each clustering with the SVM algorithm, a sub-model is built to each clustering. In the predicting process, when an incremental sample that represents new operation information is introduced in the model, the fuzzy membership function of the sample to each clustering is first computed by the FCM algorithm. Then, a corresponding SVM sub-model of the clustering with the largest fuzzy membership function is used to predict and perform incremental learning so the model can be updated on-line. An old sample chosen by heuristic sample displacement method is then discarded from the sub-model to control the size of the working set. The proposed method is applied to predict the p-xylene (PX) purity in the adsorption separation process. Simulation results indicate that the proposed method actually increases the model’s adaptive abilities to various operation conditions and improves its generalization capability.     

基于瞬时目标函数曲线特性的反应器网络综合

It is believed that whether the instantaneous objective function curves of plug-flow-reactor (PFR) and continuous-stirred-tank-reactor (CSTR) overlap or not, they have a consistent changing trend for complex reactions(steady state, isothermal and constant volume). As a result of the relation of the objective functions (selectivity or yield) to the instantaneous objective functions (instantaneous selectivity or instantaneous reaction rate), the optimal reactor network configuration can be determined according to the changing trend of the instantaneous objective function curves. Further, a recent partition strategy for the reactor network synthesis based on the instantaneous objective function characteristic curves is proposed by extending the attainable region partition strategy from the concentration space to the instantaneous objective function-unreacted fraction of key reactant space. In this paper,the instantaneous objective function is closed to be the instantaneous selectivity and several samples axe examined to illustrate the proposed method. The comparison with the previous work indicates it is a very convenient and practical systematic tool of the reactor network synthesis and seems also promising for overcoming the dimension limit of the attainable region partition strategy in the concentration space.     

On-line estimation of vapour pressure of stabilized gasoline via ANN’s

The use of artificial neural network based model for the on-line estimation of the Reid Va-por Pressure of stabilized gasoline in a stabilizer after the stripper-reabsorber in the fluid catalyticcracking unit is investigated.The quadratic basis function network(QBFN)which uses a simplequadratic function instead of sigmoid function typically used in back-propagation network is em-ployed.180 sets of historical operation data have been selected for training and testing the QBFN.To overcome the local minimum point which often occurs during the training phase,a new algorithmcombining the simulated annealing approach with the improved GDR has been applied.Furthermore,the developed model has been installed on-line in a refinery for on-line testing.Thetesting results show that the model is sufficiently accurate and it can be used on site as an on-lineanalyzer.     

一种用于计算机辅助产物设计中基于BP-NN优化问题的求解方法

Because of the powerful mapping ability, back propagation neural network (BP-NN) has been employed in computer-aided product design (CAPD) to establish the property prediction model. The backward problem in CAPD is to search for the appropriate structure or composition of the product with desired property, which is an optimization problem. In this paper, a global optimization method of using the a BB algorithm to solve the backward problem is presented. In particular, a convex lower bounding function is constructed for the objective function formulated with BP-NN model, and the calculation of the key parameter a is implemented by recurring to the interval Hessian matrix of the objective function. Two case studies involving the design of dopamine β-hydroxylase (DβH) inhibitors and linear low density polyethylene (LLDPE) nano composites are investigated using the proposed method.     

重力热管振荡传热特性RBF神经网络动态建模

The work address the problem of modeling the dynamical oscillating behavior during both unstable and stable operations, of an experimental thermosyphon. A standard RBF artificial neural network-based prediction model was developed for predicting the oscillating heat transfer of thermosyphon by means of input-output experimental measurements with the characteristics of time series. A comparison of prediction values between the RBF network and the MLP network was giving. The precision of RBF network was higher than that of the other neural networks such as BP-MLP network etc. The dynamical model of RBF network could be used to describe, predict and control the heat transfer process of a thermosyphon or a heat pipe system.     

催化重整过程的多目标优化

In this article, a multiobjective optimization strategy for an industrial naphtha continuous catalytic reforming process that aims to obtain aromatic products is proposed. The process model is based on a 20-lumped kinetics reaction network and has been proved to be quite effective in terms of industrial application. The primary objectives include maximization of yield of the aromatics and minimization of the yield of heavy aromatics. Four reactor inlet temperatures, reaction pressure, and hydrogen-to-oil molar ratio are selected as the decision variables. A genetic algorithm, which is proposed by the authors and named as the neighborhood and archived genetic algorithm （NAGA）, is applied to solve this multiobjective optimization problem. The relations between each decision variable and the two objectives are also proposed and used for choosing a suitable solution from the obtained Pareto set.     

基于粗糙集和RBF神经网络的风电场短期风速预测模型

结合粗糙集提出了一种RBF神经网络短期风速预测模型。采用粗糙集对预测模型的输入特征空间进行约简,找出对未来预测的风速具有主要影响的因素,以此作为RBF神经网络预测模型的输入变量;在RBF神经网络训练的过程中,采用在线滚动优化策略,将最新的样本加入训练集,从而使预测模型能够跟踪风速的最新变化。将提出的方法用于某风电场的1 h短期风速预测,仿真实验结果表明该方法具有结构简单、预测精度高的优点。     

A new algorithm for developing dynamic radial basis function neural network models based on genetic algorithms

A new method for extracting valuable process information from input–output data is presented in this paper. The proposed methodology produces dynamical radial basis function (RBF) neural network models based on a specially designed genetic algorithm (GA), which is used to auto-configure the structure of the networks and obtain the model parameters. The new RBF network training technique formulates a complete optimization problem, which includes the network structure into the set of free variables that are used to minimize the prediction error. This is a different approach compared with the local search methods employed by other structure selection mechanisms, which are often trapped to local minima. Another advantage of the proposed method is that only one run of the algorithm is required to obtain the optimal network structure, in contrast to the standard RBF training techniques, where the produced model is selected by trial and error. The effectiveness of the method is illustrated through the development of dynamical models for two sets of data: simulated data from a Continuous Stirred Tank Reactor (CSTR) and true data collected from a Kamyr digester, which is a rather complicated reactor used in the pulp and paper industry.     

基于模糊RBF神经网络的乙烯装置生产能力预测

针对传统的径向基函数(RBF)神经网络隐藏层节点的不确定和初始中心敏感性、收敛速度过慢等问题,提出一种基于模糊C均值的RBF神经网络(FCM-RBF)模型,通过模糊C均值聚类(FCM)得到各聚类中心,基于误差反传的梯度下降法训练隐藏层到输出层之间的权值,克服传统RBF模型对数据中心的敏感性,优化确定RBF神经网络隐藏层的节点数,提高网络训练速度和精度。最后将其用于乙烯装置生产能力预测中,分析预测不同技术、不同规模乙烯装置生产情况,指导乙烯生产,提高生产效率,结果验证了所提出算法的有效性和实用性。     

基于互信息和自组织RBF神经网络的出水BOD软测量方法

针对污水处理过程出水生化需氧量（biochemical oxygen demand，BOD）难以实时准确测量的问题，提出了一种基于互信息和自组织RBF神经网络的软测量方法对出水BOD进行预测。首先，使用基于互信息的方法提取相关特征参量作为软测量模型的输入变量；其次，设计一种基于误差校正-敏感度分析的自组织RBF神经网络，使用改进的Levenberg-Marquardt（LM）算法对网络进行训练以提高训练速度；最后将软测量模型应用于UCI公开数据集及实际的污水处理过程，实验结果表明该软测量模型结构紧凑，训练时间相对较短，预测精度有所提高，能够对出水BOD实现快速准确预测。     

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     

注塑制品翘曲变形的数值模拟优化与预测

以电池后盖板为CAE模拟分析模型,利用正交试验设计方法,将减小制品翘曲变形量作为优化目标,得到各工艺参数对制品翘曲变形量的影响程度及最优化工艺参数组合。利用径向基函数RBF神经网络对制品翘曲量进行预测,建立了各工艺参数与制品翘曲变形之间非线性映射关系模型,并与BP神经网络进行了对比。结果表明:RBF神经网络模型,可以较准备地预测制品的翘曲变形,并且在精度、训练速度等方面优于BP网络。     

基于RBF网络模型气辅成型工艺参数的优化

针对气辅成型过程中多个工艺参数的优化配置问题,提出了气辅成型工艺参数径向基函数(RBF)网络预测模型。该模型基于均匀设计思想,以RBF网络为基础,用正交最小二乘训练法对基函数的中心、方差和RBF网络权值进行优化,提高了网络预测模型对气辅成型制品质量指标的预测可靠性。以挂式空调前面板为例,对其进行CAE仿真分析,结果表明,该优化模型可实现对制品质量指标的快速预测,为工艺参数进一步优化奠定了基础。     

基于改进聚类算法的RBF网络及其应用

RBF网络可以逼近任意连续非线性函数,且训练速度快,性能好,被广泛应用于过程建模和预测。RBF网络的一个重要因素是隐层节点的选择,隐层节点过多或过少都会影响最终网络的性能。提出一种改进的k-means聚类算法,可以自动确定最优的聚类区数,并且可使最终的聚类中心合理地分布在数据空间中。在应用RBF网络进行建模和预测时,采用该方法确定隐层节点的中心,跟用通常的聚类方法相比,可以大大减小网络规模。仿真和实际应用结果都证明该方法的有效性。     

确定影响矿井风流稳定性主要风路的神经网络方法

提出了确定影响矿井通风风流稳定性主要风路的新方法。首先构造影响矿井风流稳定性的主要风路的-RBF神经网络模型,然后用生产实际数据对神经网络模型进行训练,从而确定矿井通风系统巷道的风阻与风量之间的非线性映射关系,最后用神经网络计算巷道风阻值的变化对巷道风量的影响,通过风量变化分析确定影响矿井风流稳定性的主要风路。