用人工神经网络模型预测钢的奥氏体形成温度

根据收集的实验数据，建立了预测钢的奥氏体形成温度(Ac1和Ac3点)的反向传播人工神经网络模型．用散点图和均方误差、相对均方误差和拟合分值三个统计学指标评价模型的预测性能．人工神经网络预测Ac3和Ac1的三个统计学指标分别为23．8℃，14．6℃；2．89％，2．06％和1．8921，1．7011．散点图和统计学指标均显示：人工神经网络的预测性能优于Andrews公式．此外，用人工神经网络分析了C和Mn的含量对Ac1和Ac3温度的定量影响，计算结果显示，C和Mn含量与Ac3和Ac1点间存在非线性关系，这主要是由于钢中合金元素间存在的相互作用造成的。     

基于人工神经网络的Nd-Fe-Co-Zr-B系永磁合金磁性的预测模型

为了优化合金成分以提高纳米复相Nd-Fe-Co-Zr-B系永磁合金磁性,采用均匀设计方法设计了Nd,Co,Zr和B的4因素6水平U18(6^4)实验方案,建立了合金成分与磁性之间的人工神经网络(ANN)预测模型．利用该预测模型对Nd-Fe-B合金的成分进行了优化．同时,利用所建立的人工神经网络预测模型研究了单个元素对Nd-Fe-B合金磁性的影响规律,以及多元素间的交互作用与合金磁性间的关系．结果表明：预测结果与实测结果吻合良好,预测结果的相对误差很小,Br的相对误差在1．66％以内,(BH)m的相对误差在1．94％以内,Hcj的相对误差在7．7％以内。     

人工神经网络预测马氏体开始转变点Ms

人工神经网络技术应用于马氏体开始转变点的研究。利用人工神经网络的反向传播(BP)算法，建立了材料的合金成分与马氏体开始转变点的网络模型，从而对马氏体开始转变点温度进行分析和预测，并与经验公式相比，该网络模型具有更高的精度，从而为预测马氏体开始转变点温度提供了新颖的方法。     

用人工神经网络模型预测钢的奥氏体形成温度

根据所收集的试验数据，建立了预测钢的奥氏体形成温度Ac3和Ac1点的反向传播人工神经网络模型。用散点图和均方误差、相对均方误差和拟台分值3种统计学指标评价模型的预测性能。人工神经网络预测Ac3和Ac1的3种统计学指标分别为238℃，14．6℃；2．89％，2．06％和1．8921，1.7011。散点图和统计学指标均表明人工神经网络的预测陛能优于Andrews公式。此外．用人工神经网络分析了C和．Mn的含量对Ac3和Ac1温度的定量影响．计算结果表明。C和Mn含量与Ac3和Ac1点间存在非线性关系。     

基于统计模式识别的优化BP网络在20MnVB钢冲击韧度预测中的应用

用模式识别和人工神经网络建立了20MnVB钢的化学成分与其冲击韧度之间的PLS-BP降维网络模型,用这种方法预测了20MnVB钢的冲击韧度,并与PLS回归方程的预测结果进行了比较.结果表明:所建PLS-BP网络能较准确预测20MnVB钢的冲击韧度,这为研究20MnVB钢冲击韧度提供了一种有效的方法.此外,用训练好的PLS-BP网络模型分析了合金元素B和V的含量对20MnVB钢冲击韧度的影响.     

应用人工神经网络研究化学元素对钛合金相变点的影响

钛合金的化学元素与相变点之间具有复杂的非线性关系,人工神经网络(artificial neural network,ANN)是解决非线性映射关系的一种有效可行的方法.本工作以钛合金相变点与化学元素的关系为研究对象,建立钛合金相变点的BP神经网络预测模型,运用训练好的网络模型研究典型化学元素对相变点的影响规律,并与传统经验公式进行比较.结果发现:神经网络模型的预测结果精度较高,误差小.各合金元素对相变点的影响并不是传统经验公式表现出来的单调线性关系,而是由于各元素之间的交互作用引起的复杂非线性关系.     

用人工神经网络法预测镍基单晶高温合金的蠕变断裂寿命

根据大量镍基单晶高温合金在不同温度和应力下的蠕变断裂寿命数据，采用一种先进的人工神经网络方法建立运算模型，对合金在不同实验或运行条件下的蠕变断裂寿命进行了预测，并将测算结果与现有其它方法进行了比较．结果表明，所建网络能较准确预测第一、二、三代镍基单晶合金的蠕变断裂寿命．将正交试验分析与网络预测相结合，获得在982℃／250MPa下给定合金成分范围的各元素对其蠕变断裂寿命影响程度的排序．     

人工神经网络技术探讨碳钢,低合金钢的实海腐蚀规律

根据金属海水腐蚀的特征,用人工神经网络技术分析碳钢,低合金钢海水腐蚀数据,建立了碳钢,低合金钢的腐蚀速率与合金的成分及海水间的神经网络模型,可用于预测新钢种在其它海域中的腐蚀速率,并用所建模型分析了合金元素对腐蚀速率的影响。     

专利摘要

071001内燃机活塞用耐热铝硅铸造合金[欧洲]EP1721022,2006.11.15,M.Siegfried,KS Kolbenschmidt GmbH(德国)本专利提供一种内燃机活塞用铸造铝合金,它除了含有Si以外,还含有一种以上能形成大量金属间化合物的其它合金元素,这种合金元素可选用Cu,含量为5%左右。(周静一摘译)     

用人工神经网络预测钢的贝氏体开始转变临界冷却速度

在收集的252个钢种的连续冷却转变曲线图(CCT)的基础上，用不同的人工神经网络模型预测了钢的贝氏体开始转变临界冷却速度，与实测值比较证明，不同网络模型的预测精度不同。此外，用预测精度较高的人工神经网络模型计算了合金元素Si和B的含量对贝氏体开始转变临界冷却速度的定量影响，计算结果与试验结果相符合。     

Prediction of jominy hardness profiles of steels using artificial neural networks

Jominy hardness profiles of steels were predicted from chemical composition and austenitizing temperature using an artificial neural network. The neural network was trained using some 4000 examples, covering a wide range of steel compositions. The performance of the neural network is examined as a function of the network architecture, the number of alloying elements, and the number of data sets used for training. A well-trained network predicts the Jominy hardness profile with an average error of about 2 HRC. Special attention was devoted to the effect of boron on hardenability. A network trained using data only from boron steels produced results similar to those of a network trained using all data available. The accuracy of the predictions of the model is compared with that of an analytical model for hardenability and with that of a partial least- squares model using the same set of data.     

Prediction of martensite and austenite start temperatures of the Fe-based shape memory alloys by artificial neural networks

In this study, martensite start (Ms) and austenite start (As) temperatures of Fe-based shape memory alloys (SMAs) were predicted by using a back-propagation artificial neural network (ANN) that uses gradient descent learning algorithm. An ANN model is built, trained and tested using the test data of 85 Fe-based SMAs available in literature. The input parameters of the ANN model are weight percentages of seven elements (Fe, Mn, Si, Ni, Cr, Cu and Al) and three different treatment conditions (hot rolling, homogenizing temperature and quenching). The ANN model was found to predict the Ms and As temperature well in the range of input parameters considered. A computer program was devised in MATLAB and different ANN models were constructed with this program for prediction of As and Ms temperatures of iron-based SMAs. A comprehensive analysis of the prediction errors of Ms and As temperatures made by the ANN is presented. This study demonstrate that ANN is very efficient for predicting the Ms and As temperatures of iron-based SMAs.     

Simulation of microhardness profiles for nitrocarburized surface layers by artificial neural network

A model for prediction of the microhardness profiles for nitrocarburized steels was designed. The model is based on a feed-forward artificial neural network. The performance of the model was checked, using data from the published literature as well as authors experiments. Good correspondence between predicted from the artificial neural network (ANN) and experimental data was observed. The influences of the nitrocarburizing parameters and steel composition on the microhardness profile were studied. Using the model the microhardness profiles for some steels after different conditions of nitrocarburizing were predicted. A graphical user interface was created for the use of the model.     

Prediction of density,porosity and hardness in aluminum–copper-based composite materials using artificial neural network

The potential of using feed forward backpropagation neural network in prediction of some physical properties and hardness of aluminium–copper/silicon carbide composites synthesized by compocasting method has been studied in the present work. Two input vectors were used in the construction of proposed network; namely weight percentage of the copper and volume fraction of the reinforced particles. Density, porosity and hardness were the three outputs developed from the proposed network. Effects of addition of copper as alloying element and silicon carbide as reinforcement particles to Al–4 wt.% Mg metal matrix have been investigated by using artificial neural networks. The maximum absolute relative error for predicted values does not exceed 5.99%. Therefore, by using ANN outputs, satisfactory results can be estimated rather than measured and hence reduce testing time and cost.     

熔敷金属力学性能人工神经网络预测法的应用

在实测样本的基础上,用人工神经网络建立了熔敷金属力学性能的预测模型。该模型预测的结果同实验值之间有很好的对应关系。利用该模型研究了杂质元素S,P,O,N和合金元素C,Mn,Ti对熔敷金属低温韧性的影响,并采用正交实验的方法得出了较佳的熔敷金属化学成分。本文的研究为熔敷金属力学性能的设计和控制提供了有效的途径。     

Plastic deformation of low carbon steels

Abstract

The plastic deformation of two steels was studied by means of axisymmetric compression tests carried out in a computer driven servohydraulic machine. The tests were conducted within the temperature range of 700–1100°C for steels that have similar carbon contents, but differed in other alloying elements. Little difference in strength between the steels was found when deformation took place in the austenite phase field, but the strength of the material varied as soon as ferrite was present. The critical temperatures were evaluated by means of thermal analyses, which were conducted by inserting a thermocouple within samples of the two steels; these temperatures were found to be close to those predicted by empirical equations.     

中碳合金结构钢端淬曲线的非线性方程法模拟

利用余柏海非线性模型模拟计算了中碳合金结构钢的端淬曲线，结果表明，因为该模型未考虑合金元素对钢淬透性影响的交互作用，计算所得复合合金化的中碳钢端淬曲线与实测结果偏差较大。给出了中碳合金结构钢端淬曲线模拟过程中合金元素交互作用的数学模型，引入合金化交互当量参数(L0)，并利用该改进模型对多元合金结构钢的端淬曲线进行模拟计算，结果与文献给出的曲线吻合得较好。     

Prediction of pitting corrosion of surface treated AISI 316L stainless steel by artificial neural network

Abstract

This paper presents an artificial neural network based solution method for modelling the pitting resistance of AISI 316L stainless steel in various surface treated forms. Surface treatment is a promising technique for improving the corrosion resistance of stainless steels. In this study, cyclic polarisation tests were performed before and after surface treatment. Experimental results were modelled by the neural network. The artificial neural network model exhibited superior performance based on the fitness of the observed versus predicted data. The results showed that the predicted data from the neural network model were considerably similar to the experimental data. The model has been saved and can easily be used to predict the corrosion in different surface treatment methods.