人工神经网络在材料开发中的应用研究进展

介绍了人工神经网络(ANN)的发展历程、模型特性与分类,以及反向传播(BP)神经网络模型及其改进算法,重点论述了ANN在高分子聚合反应过程和质量控制、成型加工工艺设计与条件优化、材料使用与服役性能预测方面的应用进展,以及在辅助性能表征与分析等方面的应用研究状况,并指出了ANN在未来新材料开发中应用的发展方向和亟待解决的问题。     

人工神经元网络在甲烷催化转化反应中的研究进展

人工神经元网络(ANN)是一种通过模拟大脑处理信息的方式发展起来的数据处理技术，在石油和天然气领域中被广泛用于产量预测、甲烷物性计算、甲烷吸附与分离以及甲烷催化转化等领域。针对甲烷催化转化领域，综述了近年来ANN技术在甲烷干重整、蒸汽重整、联合重整和氧化偶联反应中的应用进展，结果表明：ANN在预测甲烷转化率、产物收率等方面具有准确性好、泛化能力强、鲁棒性好的优点，在催化工艺优化、催化剂组成优化等方面也有很好的应用，对该领域存在的问题以及未来的研究方向进行了总结和展望。     

人工神经网络在注塑制品质量优化中的应用进展

大部分复杂的塑料制品均采用注射成型生产。注塑制品质量的预测和制品质量的优化是注射成型过程中的重要步骤。人工神经网络(ANN)作为人工智能最常用的方法已经被应用到注射成型中,但是,仍存在训练成本较高、构建模型复杂等缺陷。ANN预测模型可以拟合注塑过程,并且,优化注塑制品质量。以工艺参数和过程参数作为输入数据的ANN预测模型不仅可以预测注塑制品质量,还可以结合智能优化算法优化注塑工艺参数。并且,对减少ANN预测模型训练成本的方法进行了综述。最后,总结了ANN预测模型在注塑制品优化中的进展和发展方向。     

基于人工神经元网络(ANN)实现的轮胎硫化最优化过程控制

本文讨论在轮胎硫化过程中．基于ANN实现的无模型自适应调节控制。以在线自学习决定系统的ANN学习强度方式，来替代传统的PID调节控制，使硫化过程控制能够适应模型的非线性、时变性，系统的不确定因素，环境因素。     

尤里卡沥青软化点人工智能软测量技术探索

人工神经元网络(ANN)应用于软测量是人工智能方法在石化过程中成功应用的热点,ANN不需要系统模型就能映射复杂非线性关系,特别适合炼油生产过程的建模与预测工作,研究了用ANN对尤里卡沥青软化点进行软测量的方法,和现用的吴羽公司kθ法相比,ANN方法测量精度高,具有学习能力和联想记忆能力,健壮性好,它与生产装置的DCS硬件相结合能够达到优化生产控制的目的。     

基于电性拓扑状态指数的烷烃密度定量结构-性质关系研究

研究了烷烃密度的定量结构-性质关系。以电性拓扑状态指数为结构描述符,分别用多元线性回归(MLR)和人工神经网络(ANN)建立了结构描述符和密度之间的校正模型。用留一交叉验证和外部测试集验证评价所建立MLR和ANN模型的预测能力。对于MLR模型,这两种验证的均方根相对误差分别为3.37和1.92。对于ANN模型,这两种验证的均方根相对误差为1.06和1.34。这说明建立的MLR和ANN模型都可用于预测烷烃的密度,但ANN模型优于MLR模型。     

重质油黏度的定量结构-性质关系研究

研究了重质油黏度的定量结构-性质关系。将量子化学参数和拓扑指数相结合作为结构描述符,分别用多元线性回归(MLR)和人工神经网络(ANN)建立了结构描述符和黏度之间的校正模型。用留一交叉验证法,验证、评价所建立的MLR和ANN模型的预测能力。对于MLR模型,验证的均方根相对误差为7.77,对于ANN模型,验证的均方根相对误差为7.21,说明建立的MLR和ANN模型都可用于预测重质油的黏度,但ANN模型优于MLR模型。     

基于神经网络-遗传算法优化制氢工艺水碳比

根据某炼油厂制氢车间的生产数据,用人工神经网络(ANN)的反向传播(BP)算法建立了制氢装置转化生产中的水碳比神经网络预测模型,生产数据的检验表明,ANN方法能准确地关联和预报制氢装置转化生产中的水碳比,水碳比预测平均相对误差为2．83％;该神经网络预测模型用遗传算法优化并得到了最佳制氢工艺操作条件。     

基因表达式编程在含氧有机化合物定量结构-保留关系中的应用

基因表达式编程(GEP)是一种新颖的遗传算法,是一种高度有效、稳定的随机搜索方法。采用GEP对一系列含氧有机化合物的气相色谱保留指数建立定量结构-保留关系(QSRR)的模型,并与人工神经网络(ANN)预测结果进行比较。GEP和ANN在OV-1固定相上,相关系数R分别为0.9908,0.9892;在SE-54上,相关系数R分别为0.9956,0.9891。结果表明:GEP建立的模型优于ANN,具有良好的拟合度和预测精度。     

基于神经网络的挤出吹塑中型坯尺寸预测

延续了本课题组在挤出吹塑中利用人工神经网络(ANN)预测型坯尺寸的工作,建立一个新的ANN模型。经过样本训练和检验后,模型能在一定范围内预测型坯任意位置上的尺寸(直径和厚度);与以往工作相比,相同的实验量能提供更丰富的训练样本。     

An Artificial Neural Network Model for Prediction of Drying Rates

Drying rate data were generated for training of an ANN model using a liquid diffusion model for potato slices of different thicknesses using air at different velocities, humidities and temperatures. Moisture content and temperature dependence of the liquid diffusivity as well as the heat of wetting for bound moisture were included in the diffusion model making it a highly nonlinear system. An ANN model was developed for rapid prediction of the drying rates using the Page equation fitted to the drying rate curves. The ANN model is verified to provide accurate interpolation of the drying rates and times within the ranges of parameters investigated.     

An Artificial Neural Network Model for Prediction of Drying Rates

Abstract

Drying rate data were generated for training of an ANN model using a liquid diffusion model for potato slices of different thicknesses using air at different velocities, humidities and temperatures. Moisture content and temperature dependence of the liquid diffusivity as well as the heat of wetting for bound moisture were included in the diffusion model making it a highly nonlinear system. An ANN model was developed for rapid prediction of the drying rates using the Page equation fitted to the drying rate curves. The ANN model is verified to provide accurate interpolation of the drying rates and times within the ranges of parameters investigated.     

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     

Recent Advances in Solid State NMR of Small Molecules in Confinement

In the past years, porous media have become a major focus of materials science, due to their versatile properties, such as high surface area, low specific weight, high surface functionality, and the ability to customize their surface properties. Applications of porous media range from catalysis to separation media to gas storage. All of the mentioned applications involve the introduction of guest molecules into the pores. For efficient application of the materials, it is essential to know the behavior of these introduced molecules in the confined state. Solid state (ss) NMR gives a unique insight into the dynamics, the guest-host interactions, and the binding sites of porous materials and is probably the most powerful characterization method for probing a huge variety of real-life systems. Recent results in research of microporous zeolites and periodically mesoporous silica (PMS) materials using NMR will be highlighted.     

Intelligent control system for extractive distillation columns

We developed and implemented an intelligent control system to be used in an extractive distillation column that produces anhydrous ethanol using ethylene glycol as solvent. The concept of artificial neural networks (ANN) was used to predict new setpoints after disturbances, and proved to be a fast and feasible solution. The developed control system receives data from temperature, flowrate and composition measurements of the azeotrope feed, and the ANN estimates the new set-points of the controllers to maintain 99.5 mol% of ethanol at the top and less than 0.1mol% at the bottom; feed composition was also estimated using an ANN. All ANN were trained to provide output data corresponding to an optimized operating condition. The results showed that the intelligent control system can predict a new operating condition for any disturbance in the column feed and presented superior performance when compared with the control system without ANN.     

ZrO2-SiC材料中原位SiC生成量预报的神经网络模型

研究了基于神经网络的ZrO2-SiC材料中原位SiC生成量预报模型,运用材料制备过程中的工艺参数,实现了SiC生成量的预报.结果表明:本模型具有良好的预报效果,人工神经网络是材料性能定量预报的一种有效方法.     

Dan Shechtman’s Quasicrystal Discovery in Perspective

Dan Shechtman’s discovery of quasicrystals brought about a paradigm change in chemistry, physics, materials science, and other areas of science and engineering. Although superficially it could be looked at as a serendipitous event, Shechtman’s curiosity and drive played equal parts with serendipity in this discovery. Shechtman was a lonely discoverer, again, seemingly detached from the main stream of generalized crystallography for which his contribution was a milestone. Generalized crystallography is the science of structures without restrictions — “structures beyond crystals.” 1 The discovery of quasicrystals can be seen as written into the history of ideas that have much extended our views about the tools of our scientific inquiry and the materials we aim at producing and utilizing. This review augments a recent Editorial in the August 2011 issue of Structural Chemistry about the lessons of the quasicrystal discovery 2 and a book chapter about Dan Shechtman’s traits as a discoverer and about his road to the discovery. 3     

基于人工神经网的催化裂化故障诊断研究

由于催化裂化系统的变量众多,每个故障的发生都可能涉及众多操作变量,这给故障的准确判断带来困难.人工神经网具有很强的自适应能力,非线性映射能力和泛化能力也很强,对数据进行高度并行处理,可自然处理多输入信号.用于催化裂化的故障诊断,取得较好的效果.将改进的LM算法用于人工神经网的训练,提高了网络的收敛速度.     

发酵过程生物量软测量技术的研究进展

生物量是发酵过程中的关键过程参数之一,它直接影响着发酵过程的优化和控制。综述了近年来发酵过程生物量软测量技术的研究现状,讨论了基于过程机理分析、回归分析、状态估计和神经网络等的软测量建模方法,对基于神经网络和改进的神经网络建模方法进行了分析。指出基于多尺度建立软测量混合模型,是实现发酵过程生物量在线测量的有效方法,并给出了建立混合模型需要解决的关键问题。     

Auto-analysis system for graphite morphology of grey cast iron

The current method to classify graphite morphology types of grey cast iron is based on traditional subjective observation, and it cannot be used for quantitative analysis. Since microstructures have a great effect on the mechanical properties of grey cast iron and different types have totally different characters, six types of grey cast iron are discussed and an image-processing software subsystem that performs the classification and quantitative analysis automatically based on a kind of composed feature vector and artificial neural network (ANN) is described. There are three kinds of texture features: fractal dimension, roughness and two-dimension autoregression, which are used as an extracted feature input vector of ANN classifier. Compared with using only one, the checkout correct precision increased greatly. On the other hand, to achieve the quantitative analysis and show the different types clearly, the region segmentation idea was applied to the system. The percentages of the regions with different type are reported correctly. Furthermore, this paper tentatively introduces a new empirical method to decide the number of ANN hidden nodes, which are usually considered as a difficulty in ANN structure decision. It was found that the optimum hidden node number of the experimental data was the same as that obtained using the new method.