Heat effects due to mixing (dilution) the mixed acid solutions—Application of neural networks to approximate and generalize experimental data

The feed-forward neural networks have been used to approximate the specific molar enthalpy and the specific molar heat capacity of the mixed acid solutions. The nets have been trained with experimental data taken from the literature, so the values of the specific molar enthalpy and the specific molar heat capacity at the reference temperature T = 0 °C could be successively estimated for any composition of the mixed acid. Two principal methods have been considered and tested. In the first method two independent neural nets have been employed: the net NN-H, which approximates separately the specific molar enthalpy and the net NN-C, to approximate the specific molar heat capacity, respectively. In the second method only one net is employed (the net NN-HC), which simultaneously approximates both the specific molar enthalpy and the specific molar heat capacity. Then following both mentioned methods, the trained neural nets have been used to model the heat effects due to dilution of mixed acid solutions, carried out at various conditions – i.e. at any temperature and composition. Using these nets, both, the integral and the differential enthalpy balance can be carried out, so the smart and accurate method to model the mixed acid dilution has been elaborated. The proposed methods and their prediction accuracy have been successfully verified with our own experimental data carried out in the RC1 reaction calorimeter.     

神经网络在离心式压缩机喘振监测中的应用

简述了神经网络理论和几种改进 B P 网络的原理及特点,重点介绍了尺度共轭梯度算法。把尺度共轭梯度神经网络用于机组工艺量,用于对压缩机喘振进行监测,提出了对喘振预防的一种更精确的全新方法。     

自制微孔曝气装置在焦化生物脱酚系统上的应用

针对焦化废水生物脱酚系统溶解氧过低的问题,通过淘汰原有竖管曝气装置,采用自制微孔曝气装置,使曝气量完全达到工艺要求,同时污泥沉降比得到提高,相比成熟的尼龙网微孔曝气装置,节约成本,便于维护。     

径向基函数神经网络在砂体厚度预测中的应用

目前,人工智能神经网络在地震储层参数的预测方面具有广泛的应用,最常用的为BP神经网络,但是效果并不是十分理想。径向基函数神经网络(RBFN)是一种前馈神经网络,其在函数逼近、模式识别方面都优于BP网络,已经在岩性识别、孔渗预测方面取得了较好的应用效果。本文首次将此方法运用于预测砂体厚度,利用地震属性信息和神经网络的学习,基于实际数据计算,最后计算出相应的砂体厚度值,并与实测值进行误差分析。实例分析表明,利用径向基函数神经网络进行砂体厚度预测具有一定的可行性和实用价值。     

An Extracellular Perspective on CNS Maturation: Perineuronal Nets and the Control of Plasticity

During restricted time windows of postnatal life, called critical periods, neural circuits are highly plastic and are shaped by environmental stimuli. In several mammalian brain areas, from the cerebral cortex to the hippocampus and amygdala, the closure of the critical period is dependent on the formation of perineuronal nets. Perineuronal nets are a condensed form of an extracellular matrix, which surrounds the soma and proximal dendrites of subsets of neurons, enwrapping synaptic terminals. Experimentally disrupting perineuronal nets in adult animals induces the reactivation of critical period plasticity, pointing to a role of the perineuronal net as a molecular brake on plasticity as the critical period closes. Interestingly, in the adult brain, the expression of perineuronal nets is remarkably dynamic, changing its plasticity-associated conditions, including memory processes. In this review, we aimed to address how perineuronal nets contribute to the maturation of brain circuits and the regulation of adult brain plasticity and memory processes in physiological and pathological conditions.     

基于径向基神经网络的聚丙烯熔融指数预报

引　言在化工生产中大部分生产流程具有非线性、大时滞、结构复杂等特性 ,而且生产变量之间存在着不同程度的耦合与关联 .前馈神经网络由于具有强大的拟合非线性函数的能力 ,已成为生产指标预测的有力工具[1] .其中径向基 (radialbasisfunction ,RBF)神经网络相对于神经网络BP     

Application of Different Artificial Neural Networks Retention Models for Multi-Criteria Decision-Making Optimization in Gradient Ion Chromatography

In this work, the principles of multi-criteria decision-making were used to develop an efficient optimization strategy in gradient elution ion chromatographic analysis. Two different artificial neural network retention models (multi-layer perceptron and radial basis function), three different separation criterion functions (chromatography response function, separation factor product and normalized retention difference product), and four different robustness criterion functions (CR1-CR4) were examined. The shape of the calculated separation vs the robustness response surface was used as principal criterion. Analysis time and minimum separation of adjacent peaks were additional criteria. The results showed that the radial basis artificial neural network retention model in combination with normalized retention difference product separation criterion function and CR3 robustness criterion function provided the optimal gradient ion chromatographic analysis.     

Simulation of Temperature Distribution in Stored Wheat without Aeration

A neural-deterministic simulation model applied for calculating distributions of temperatures and moisture content in a bed of wheat stored in a steel silo without aeration is presented in the article. The model consists of differential equations of heat and moisture transfer, initial and boundary conditions, and three artificial neural networks used during simulated ambient air conditions. Experiments and computer simulations were carried out in order to determine temperature fields in wheat grain stored in a steel silo for two months. The computer simulations were carried using MATLAB and FEMLAB software. The difference between measured and simulated temperature in grain near the silo wall at a height of 2.5 m from the bottom was less than 3.0°C. On the basis of the analysis performed it was concluded that the temperature distributions obtained with the model were consistent with the measured results obtained for grain stored in a steel silo without aeration.     

Simulation of Temperature Distribution in Stored Wheat without Aeration

A neural-deterministic simulation model applied for calculating distributions of temperatures and moisture content in a bed of wheat stored in a steel silo without aeration is presented in the article. The model consists of differential equations of heat and moisture transfer, initial and boundary conditions, and three artificial neural networks used during simulated ambient air conditions. Experiments and computer simulations were carried out in order to determine temperature fields in wheat grain stored in a steel silo for two months. The computer simulations were carried using MATLAB and FEMLAB software. The difference between measured and simulated temperature in grain near the silo wall at a height of 2.5 m from the bottom was less than 3.0°C. On the basis of the analysis performed it was concluded that the temperature distributions obtained with the model were consistent with the measured results obtained for grain stored in a steel silo without aeration.     

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RBF神经网络的油藏相对渗透率曲线计算

提出了一种基于改进的RBF神经网络的相对渗透率曲线计算方法。利用骨干粒子群的位置更新操作更新RNA遗传算法的变异算子得到混合RNA遗传算法（HRGA）,针对RBF神经网络中隐含层径向基中心值的确定,利用HRGA算法对其进行优化,并用于相对渗透率曲线的计算。将HRGA优化的RBF神经网络和标准RBF神经网络计算的相对渗透率曲线与真实值误差对比分析,实验结果表明HRGA优化的RBF神经网络明显提高了计算精度。     

基于迭代启发式动态规划的锅炉燃烧系统的优化控制

对锅炉燃烧系统进行径向基函数(RBF)神经网络建模,利用基于RBF神经网络的迭代启发式动态规划(HDP)算法对锅炉燃烧系统进行优化控制,并对神经网络初始权值和效用函数进行改进后与传统的HDP算法进行比较分析.     

Hybrid neural network‐based prediction model for tribological properties of polyamide6‐based friction materials

The hybrid neural network employed to predict the tribological properties of solid lubricants reinforced nano‐TiO₂/polyamide6 composites was established based on the back propagation and radial basis function networks, and optimized by adaptive genetic algorithm. With such three factors as material composition, testing loads, and velocities, orthogonal tests were designed and the data obtained was used for training the neural network. The correlation index between the predicted and the experimental values for friction coefficient and ware rate were 0.992 and 0.998, respectively, and 3D plots for the predicted friction coefficient and wear rate as a function of material compositions and testing conditions were established. It shows that the results are in good agreement with the data measured. It is demonstrated that the well‐optimized neural network had remarkable capability for modeling concern. POLYM. COMPOS., 38:1705–1711, 2017. © 2015 Society of Plastics Engineers     

薄管板结构强度计算的新思路

介绍了薄管板的概念及结构特点 ,叙述了薄管板结构强度分析及应用RBF(RadialBasisFunc tions径向基函数 )神经网络对薄管板结构的管板应力计算进行系统辨识的过程 ,从而提出了一种用多个具有一定意义的无量纲参数函数式表达薄管板结构强度计算的新思路 ,并说明在结构强度分析中应用神经网络方法是实用的     

基于神经网络的两相流流型识别方法研究

采集了水平管内气水两相流动的差压信号,利用概率密度函数(PDF)对差压信号特征进行了分析;定义了PDF的四个特征参数,即PDF波峰个数K1、波峰峰值K2、波峰位置K3、以及PDF的方差K4来反映流型的特征。运用四个参数构成的特征向量对径向基函数(RBF)神经网络进行训练并识别流型,结果表明,该方法具有识别速度快、准确率高的特点,从而为两相流的流型识别提供了一种有效的手段。     

Use of neural net models for statistical analysis and regulation of glass ribbon formation on tin melt

A method for constructing and applying neural-net models for analyzing and regulating the conditions for glass ribbon formation on tin melt is examined. In this method models are constructed on the basis of neural nets describing the dependence of the indicators of glass properties on the regime and controllable variable input parameters. An algorithm for adjusting the regime variables as a function of the thickness of the glass ribbon produced is developed taking account of changes occurring in the effects which are under observation.     

运用RBF神经网络设计复合材料界面性能

提出一种设计碳纤维织物/环氧复合材料界面性能的新方法。利用径向基函数神经网络，建立起工艺参数与复合材料界面性能的关系模型。同时给出实例来验证此方法的有效性。     

基于RBF神经网络的流向变换催化燃烧反应器的温度预测

用RBF神经网络建立了用于清除低浓度挥发性有机物的流向变换催化燃烧反应器拟定态温度分布模型.从基于过程机理模型的数值计算结果出发,结合中试装置的实时操作数据建立了拟定态床层温度的人工神经元网络深层知识库,用于增强神经网络模型的“外推能力”和“可信度”.仿真结果表明所建立的模型简单、精度高,能满足特性预测的要求.     

Comparative performance analysis of spectral estimation algorithms and computational optimization of a multispectral imaging system for print inspection

We have analyzed the performance of simulated multispectral systems for the spectral recovery of reflectance of printer inks from camera responses, including noise. To estimate reflectance we compared the performance of four algorithms which were not comparatively tested using the same data sets before. The criteria for selection of the algorithms were robustness against noise, amount of data needed as inputs (training set, spectral responsivities) and lacking of use of dimensionality reduction techniques. Three different sensor sets and training sets were used. We analyzed the differences in the spanning of the subspaces found for the three training sets, concluding that the ink reflectances have characteristic features. The best performance was obtained using the kernel and the radial basis function neural‐net‐based algorithms for the training set composed of printer inks reflectances, whereas for the other two training sets (composed of samples from the ColorChecker DC and Vhrel's reflectances' set) the quality of the recovered samples was more uniform among the algorithms. We also have performed an optimization to choose the best sensor set for the multispectral system with a reduced number of sensors. © 2012 Wiley Periodicals, Inc. Col Res Appl, 39, 16–27, 2014