基于BP神经网络的结晶成核速率预测

利用神经网络所具有的输入输出之间的高度非线性映射关系,给出了一种利用BP神经网络模型预测磷酸二氢铵结晶成核速率的方法。在对网络进行训练的基础上,建立了磷酸二氢铵结晶生长速率与过饱和度、冷却温度、饱和温度及悬浮密度和之间的数学模型。仿真结果表明,利用文中所提出的神经网络模型能够较准确、快速地预测结晶成核速率的变化,预测值与测量值的最大相对误差不超过5.9%,表明该网络预测模型有很大的实用性。     

Analysis of the amino acid effect on protein folding by atom pair contacts

New atomic pair contacts with considering the coordinates of each atom in a residue are introduced here. We analyze the ability of all the 20 amino acid residues to form long-range and short-range contacts by calculating the average numbers of short- and long-range contacts between different amino acid pairs. It is concluded that Phe-Phe, Leu-Phe and Leu-Leu have a high tendency to form contacts. The relative ability to form atom pair contact does not depend on the limiting value of R_C. The average number of contacts per residue, which is the scale of the relative ability to form contacts for the 20 amino acid residue types, is also calculated. The result shows that hydrophobic residues with large numbers of long-range contacts more easily form long-range contacts, while the hydrophilic ones form long-range contacts less often. Linear regression analysis by a new method of counting contacts concludes that either contact order (CO) or total contact distance (TCD) parameter has a significant correlation with the logarithms of folding rates. The relative deviations between the experimentally observed and the two parameters CO and TCD are smaller than that with previous methods. Moreover, the values of CO_λ-μ and TCD_λ-μ between λ-type and μ-type amino acids are investigated. Comparisons between the Fauchere-Pliska hydrophobicity scale and the average number of contacts per residue formed are also made. The new knowledge of atomic pair contacts can help us understand the importance of amino acid residue type and its sequence in globular structure of the protein in detail.     

煤种成浆性的人工神经网络预测模型研究

为更好地预测煤的成浆性,以大量煤种成浆浓度试验数据为基础,建立了3个输出因子的神经网络成浆浓度预测模型,模型采用L-M算法,对输入数据进行数据预处理,最后对比分析了神经网络预测模型与回归分析模型的预测结果。结果表明,以A_d、哈氏可磨性指数HGI和氧含量O为输入因子的模型预测结果平均绝对误差为0.63%,以M_(ad)、HGI和O为输入因子的模型预测结果平均绝对误差为0.60%,以M_(ad)、HGI和氧碳比O/C为输入因子的模型预测结果平均绝对误差为0.40%,3种组合的模型结果均小于回归分析模型的平均绝对误差1.15%。因此神经网络模型比回归分析模型有更好的预测能力,其中以M_(ad)、HGI和O/C为输入因子的神经网络模型预测结果最好。     

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。     

基于RBF网络的胶磷矿浮选精矿指标预测模型

本文基于RBF神经网络构造了云南某胶磷矿浮选多因素输入和浮选精矿品位、回收率之间的浮选模型,并在Matlab环境下进行了计算机仿真试验,结果表明,模型预测精度较高,验证了非参数建模的合理性,具有一定的实用价值,为浮选过程的控制奠定了基础.     

An analysis for effect of cetane number on exhaust emissions from engine with the neural network

Decoupling cetane number from the other compositions and properties of diesel fuel, the individual effect of cetane number on the exhaust emissions from an engine may be researched. This paper has presented a back-propagation neural network model predicting the exhaust emissions from an engine with the inputs of total cetane number, base cetane number and cetane improver, total cetane number and nitrogen content in the diesel fuel; as well as the output of the exhaust emissions of hydrocarbon (HC), carbon oxide (CO), particulate matter (PM) and nitrogen oxide (NO_x). An optimal design has been completed for the number of hidden layers, the number of hidden neurons, the activation function, and the goal errors, along with the initial weights and biases in the back-propagation neural network model. HC, CO, PM and NO_x have been predicted with the model, the effects of cetane improver and nitrogen content on them have also been analyzed, and better results have been achieved.     

基于径向基函数神经网络的挤出温度预测模型

简要介绍了软测量方法建立挤出温度预测模型的方法,分析其优缺点后进而提出应用基于RBF神经网络建立渝度预测模型,试验后对比了实测值和预测值,结果表明该方法能达到较好的预测精度,同时具有使用简洁、快速等优点,具有较好的应用推广意义。     

基于BP神经网络电子评标方法的研究和应用

为了解决招投标中评标环节专家评审法随意性大的问题,针对建设工程的不同特点,对评标内容进行分类剖析.利用人工神经网络的智能识别功能,建立基于神经网络的电子评标方法.结合人工神经网络的自适应功能、良好的容错能力以及非线性数据处理特点,科学地评价多因素建设工程投标方案,使招标商从众多的投标商中选出最佳的中标单位.     

Dynamic prediction of milk ultrafiltration performance: A neural network approach

Neural network models were tested in connection with the dynamic prediction of permeate flux (J_P), total hydraulic resistance (R_T) and the solutes rejection for the crossflow ultrafiltration of milk at different transmembrane pressure (TMP) and temperature (T). This process has complex non-linear dependencies on the operating conditions. Thus it provides demanding test of the neural network approach to the process variables prediction. Two neural network models with single hidden layer were constructed to predict the time dependent rate of J_P/R_T and rejections from a limited number of training data. The modelling results showed that there is an excellent agreement between the experimental data and predicted values, with average errors less than 1%. The experimental results showed that the R_T and solutes rejection (except for protein) increased greatly with time at each value of TMP and T, whereas the J_P decreased significantly for the same conditions. Increasing TMP at constant T led to an increase in the J_P, R_T and solutes rejection, but increasing T at constant TMP had no significant effect on the J_P, R_T and rejection of components.     

GA-BP神经网络的钛合金微弧氧化膜层厚度预测模型的建立

利用正交试验法获得的TC4钛合金微弧氧化实验数据建立了基于4-11-1(即4个输入神经元,11个隐含层节点,1个输出神经元)结构的BP神经网络预测膜层厚度的模型,并引入遗传算法(GA)对其权值和阈值进行优化。以微弧氧化工艺参数中的电流密度、脉冲频率、占空比和氧化时间作为网络的输入向量,氧化膜层厚度作为网络的输出向量,对比和分析了BP与GA-BP模型的预测结果。与BP网络模型相比,GA-BP网络模型稳定性能较好,并能高精度预测膜层的厚度,GA-BP网络模型预测值的平均误差为0.015,最大误差仅为0.036,而BP模型预测结果的平均误差为0.064,最大误差为0.099。     

Differential recurrent neural network based predictive control

In this paper an efficient algorithm to train general differential recurrent neural network (DRNN) is developed. The trained network can be directly used in the nonlinear model predictive control (NMPC) context. The neural network is represented in a general nonlinear state-space form and used to predict the future dynamic behavior of the nonlinear process in real time. In the new training algorithms, the ODEs of the model and the dynamic sensitivity are solved simultaneously using Taylor series expansion and automatic differentiation (AD) techniques. The same approach is also used to solve the online optimization problem in the predictive controller. The efficiency and effectiveness of the DRNN training algorithm and the NMPC approach are demonstrated through a two-CSTR case study. A good model fitting for the nonlinear plant at different sampling rates is obtained using the new method. A comparison with other approaches shows that the new algorithm can considerably reduce network training time and improve solution accuracy. The DRNN based NMPC approach results in good control performance under different operating conditions.     

Artificial neural network approaches on composition-property relationships of jet fuels based on GC-MS

The relationships of composition-properties of 80 jet fuels concerning chemical compositions and several specification properties including density, flashpoint, freezing point, aniline point and net heat of combustion were studied. The chemical compositions of the jet fuels were determined by GC-MS, and grouped into eight classes of hydrocarbon compounds, including n-paraffins, isoparaffins, monocyclopraffins, dicyclopraffins, alkylbenzens, naphthalenes, tetralins, hydroaromatics. Several quantitative composition-property relationships were developed with three artificial neural network (ANN) approaches, including single-layer feedforward neural network (SLFNN), multiple layer feedforward neural network (MLFNN) and general regressed neural network (GRNN). It was found that SLFNNs are adequate to predict density, freezing point and net heat of combustion, while MLFNNs produce better results as far as the flash point and aniline point prediction are concerned. Comparisons with the multiple linear regression (MLR) correlations reported and the standard ASTM methods showed that ANN approaches of composition-property relationships are significant improvement on MLR correlations, and are comparable to the standard ASTM methods.     

Development of an artificial neural network model for prediction of cell voltage and current efficiency in a chlor-alkali membrane cell

This paper presents the development of an artificial neural network (ANN) model for the prediction of cell voltage and caustic current efficiency (CCE) versus various operating parameters in a lab scale chlor-alkali membrane cell. In order to validate the model predictions, the effects of various operating parameters on the cell voltage and current efficiency of the membrane cell were experimentally studied. The membrane cell incorporated a standard DSA/Cl₂ electrode as the anode, a nickel electrode as the cathode and a Flemion 892 polymer film as the membrane. Each of the six process parameters including anolyte pH (2–5), operating temperature (25–90 °C), electrolyte velocity (2.2–5.9 cm/s), brine concentration (200–300 g/L), current density (1–4 kA/m²), and run time were thoroughly studied at four levels and low caustic concentrations (5–22 g/L). The predictions of ANN model as well as those from other statistical methods were evaluated versus the measured values of cell voltages.

The developed ANN model is not only capable to predict the cell voltage and caustic current efficiency but also to reflect the impacts of process parameters on the same functions. The predicted cell voltages and current efficiencies using ANN modeling were found to be close to the measured values, particularly at higher current densities.     

基于BP神经网络的煤与瓦斯突出危险性的预测研究

应用BP神经网络的理论和方法,结合贵州某矿山的实际情况,建立了基于BP神经网络的煤与瓦斯突出危险性预测的数学模型,通过数学软件matlab 7.0对煤层的突出危险程度进行了预测。结果显示,煤与瓦斯突出危险性的预测与实际情况相符,表明采用BP神经网络模型进行预测是可行的,为煤与瓦斯突出的预测提供了一种精度较高的方法。     

基于PSO-BP-PID神经网络的注塑机料筒温度预测算法研究

提出了一种基于粒子群优化(PSO)算法、BP神经网络及比例积分微分(PID)控制的复合算法的注塑机料筒温度预测模型,即PSO-BP-PID神经网络模型,并进行了仿真研究.结果表明:使用PSO算法确定该模型的输出权重,并且对混合核函数参数进行优化升级;在模型训练过程中,使用更大的容许度处理正误差,保证预测误差始终处于正值...     

BP神经网络在合成镁钙砂中的应用

利用人工神经网络技术 ,建立BP网络模型 ,通过网络的学习训练 ,能够比较准确地预测影响合成镁钙砂粉化率的因素     

基于神经网络控制的共轭梯度法

共轭梯度法中搜索步长是通过某种搜索策略得到,许多情况下的收敛速度较慢。为了加快其收敛速度,提出了通过引入具有“先验知识”的神经网络对共轭梯度算法中的搜索步长进行控制。实验结果表明,该模型实现的共轭梯度法对于加快收敛速度有效。     

基于小波神经网络的污水处理污泥体积指数软测量

污水处理过程具有非线性、时变和滞后等特点,因而无法进行准确的数学建模。现有的污水处理技术中最突出的问题是一些关键的水质参数不能在线监测,只能通过人工间接测量再通过计算获得,耗时较长,不能及时地进行信息反馈,会造成一些严重的后果。为了避免这样的问题,提出了基于小波分析的神经网络(BP)软测量技术,通过建立小波神经网络参数软测量模型,对污水处理中难测水质参数SVI(污泥体积指数)进行在线监测。研究表明,此方法能有效规避单一的BP算法收敛速度慢、容易陷入局部最优解等问题,有助于实现对污水处理的智能控制。     

BP神经网络计算乙醇-环己烷-水体系汽-液平衡

基于带动量因子的 BP神经网络 ,以实验测定的乙醇 (1) -环己烷 (2 ) -水 (3)体系在 35℃、5 0℃、6 5℃的汽液平衡数据为训练和预测样本进行了计算 ,选择温度、X1 和 X2 3个参数作为输入 ,Y1 、Y2 和 Y3作为输出 ,隐层单元数为 9,学习速率为 0 .5 ,动量因子为 0 .12 8。对 Y1 ,Y2 ,Y3,神经网络计算的训练平均误差分别为 :0 .0 0 71,0 .0 101,0 .0 0 6 0 ,预测平均误差分别为 :0 .0 0 6 5 ,0 .0 12 4 ,0 .0 0 6 0 ,小于 NRTL 模型计算的相应误差。为相平衡计算提供了新的有效的工具。     

矿井通风系统评价的人工神经网络模型

建立了矿井通风系统评价的人工神经网络模型。在１３个金属矿山的实际应用表明，该评价方法简单、实用。