基于BP算法的Ti-Fe-Mo-Mn-Nb-Zr系钛合金成分优化

研究了不同成分的Ti Fe Mo Mn Nb Zr合金的压缩强度、硬度及其在Hank’s人工体液中的耐腐蚀性能 ,得到了高压缩强度、硬度与牙本质相近的钛合金。在综合考虑钛合金成本、力学性能及耐腐蚀性能的基础上 ,采用BP神经网络建立了钛合金中的元素质量分数与硬度之间的关系 ,并通过实验进行了验证 ,预测结果与实验测定结果的对比是令人满意的。在保证性能的基础上 ,利用训练好的BP网络结构对材料的成分进行优化 ,以减少贵重元素的加入量。     

Modeling of mechanical properties of as-cast Mg-Li-Al alloys based on PSO-BP algorithm

Artificial neural networks have been widely used to predict the mechanical properties of alloys in material research.This study aims to investigate the implicit relationship between the compositions an...     

铝电解质过热度预测模型研究

根据在线式与离线式测量初晶温度的特点,提出了一种基于Rerstreken公式改进的离线式初晶温度预测模型,进而根据初晶温度计算得出铝电解质的过热度。模型以铝电解质中各种成分的质量百分数和取样分析周期的电解数据作为辅助变量,考虑取样分析周期内A lF3浓度的变化,实现了对初晶温度的准确估计。试验结果表明,本文所提出的预测模型能够准确地预测铝电解质的过热度。     

A fast algorithm for strain prediction in tube hydroforming based on one-step inverse approach

This paper presents recent developments of a simplified finite element method called the inverse approach (IA) for the estimation of large elastoplastic strains and thickness distribution in tube hydroforming. The basic formulation of the IA, proposed by Guo et al. (1990), has been modified and adapted for the modeling of three-dimensional tube hydroforming problems in which the initial geometry is a circular tube expanded by internal pressure and submitted to axial feed at the tube ends. The application of the IA is illustrated through the analyses of numerical applications concerning the hydroforming of axisymmetric bulge, made from aluminum alloy 6061-T6 tubing, the hydroforming of square section hollow component and the hydroforming of a free Tee extrusion from welded low carbon steel LCS-1008 tubing. Verifications of the obtained results have been carried out using experimental results together with the classical explicit dynamic incremental approach using ABAQUS^® commercial code to show the effectiveness of our approach.     

基于改进的最小二乘支持向量机的退火炉内带钢延伸量预测

带钢在退火炉内进行长时间加热的过程中,由于炉内温度、张力以及带钢运行速度等变量间具有多重相关性且存在生产数据噪声,从而难以准确预测炉内带钢的延伸量。本文提出了一种基于遗传算法的BP神经网络理论(GA-BP神经网络)与最小二乘支持向量机(LSSVM)理论相结合的算法用于目标函数拟合,该算法兼顾最小二乘支持向量机的全局与局部拟合能力,有效避免算法陷入局部最优的缺点,利用改进的BP神经网络算法优化选择LSSVM 模型的惩罚因子和核函数参数,以避免人为选择参数的盲目性,使LSSVM具有更好的泛化能力和预测精度。将该方法应用到退火炉内带钢的延伸量预测,通过现场生产数据仿真模拟进行非线性函数估计,结果表明：本文所提出的方法具有良好的预测精度,可以应用于实际生产。     

Ti-Cr合金的非等温氧化及着火预测(英文)

采用热重分析、XRD和SEM等方法研究Ti-Cr合金(0≤w(Cr)≤25%)从室温至1723K的非等温氧化行为及氧化膜的微观结构,探讨Cr元素对Ti-Cr合金抗氧化能力的影响机制。结果表明:当Cr含量小于某一临界值wC时,随着Cr含量的增加合金的抗氧化能力降低;当Cr含量高于wC时,随着Cr含量的增加合金的抗氧化能力提高;当温度高到1000K时,Ti-Cr合金的氧化仍符合抛物线规律,且主要发生钛的氧化;Ti-Cr合金氧化后基体中存在氧扩散层,氧化膜主要为金红石型TiO2,内层氧化膜出现富Cr现象,Cr氧化物的析出提高了Ti-Cr合金的抗氧化能力。金属和合金的着火是一个快速非等温氧化的过程,预测了Ti-Cr合金着火阶段的氧化机制。     

基于遗传算法优化BP神经网络的内螺纹冷挤压质量预测

为了预测不同工艺参数下的挤压扭矩、挤压温度及螺纹成形质量,基于MATLAB搭建了BP-GA神经网络预测模型,利用遗传算法对BP神经网络权值和阈值进行优化,使预测结果更加精确.结果 表明:BP-GA神经网络预测模型对内螺纹冷挤压过程中的挤压扭矩、温度和牙高率的预测精度较高,挤压扭矩的试验值与预测值误差为10％～15％,挤...     

基于功率最小原理和遗传算法的钢轨万能轧制过程宽展预测

为得到钢轨万能轧制过程轨头和轨底的精确宽展,以轨头和轨底宽展为待定参数,在坯料断面尺寸、设备参数、轧制工艺参数已知的基础上,基于上限原理推导得出包含轨头和轨底宽展系数的钢轨万能轧制过程总功率表达式,并应用MATLAB软件遗传算法工具箱编写M文件对钢轨万能轧制总功率进行单目标优化,得到轧制功率最小时的轨头和轨底宽展系数。为验证理论模型,对18kg/m轻轨进行万能轧制实验,并用DEFORM软件模拟其万能轧制过程,将优化后宽展系数与实验结果进行对比,两者误差小于2%。     

Thermodynamic prediction of the composition of Fe-based amorphous alloys

The liquidus univariant lines of the Fe-Nb-B ternary system have been thermodynamically calculated by means of CALPHAD method and Fe-based thermodynamic data. It is found that there are two eutectic reactions in the Fe-rich corner,that is,（1） L（Fe-3Nb-15B） →α＋γ＋ M2B （1430 K）,and （2） L（Fe-10Nb-27B） → FeB ＋ Lc14 ＋ M2B （1575 K）. Moreover,the eutectic points are very close to the compositions with high glass forming ability determined experimentally. This means that it is feasible to design the compositions of multicomponent bulk metallic glasses by looking for the eutectic points in the Fe-Nb-B system by means of thermodynamic calculation.     

基于遗传神经网络的镍基高温合金激光熔覆层形貌质量预测

采用反向传播(back propagation,BP)人工神经网络(artificial aeural network,ANN)和遗传算法建立了激光熔覆层形貌质量(熔覆层高度、宽度及稀释率)与激光功率、送粉速率和扫描速率之间的遗传神经网络预测模型.设计正交试验得到预测模型训练样本数据,并在正交试验的基础上,用极差分析法分析了各加工参数对熔覆层形貌质量各个指标的影响规律.经过试验验证,遗传神经网络模型预测值与试验实测值误差不大于4.6%.结果表明,运用该模型可以为准确的选择镍基高温合金激光熔覆参数提供一定参考,从而有利于提高镍基高温合金激光熔覆层形貌质量.     

Molten aluminium attack on iron based alloys

Liquid aluminium alloys affect the moulds, by formation of intermetallic layers at their surface; such attack is enhanced by the flow of the molten metal on the surface of the tooling. Metal tooling are used to promote high heat transfer rates and increase the solidification rate to enhance mechanical properties in cast pieces, but the damage to their surface affects their quality and increases the production costs due to the need to repair them. This work presents the results of a series of studies conducted to evaluate the attack produced by molten aluminium on samples of iron based materials commonly used in the manufacture of dies and moulds. The studies were conducted in a machine that imposes a rotating movement to experimental specimens to simulate the flow of liquid metal. The damage observed in the samples was compared to that experienced by tooling in the melt shop.     

Intermetallic alloys based on gamma titanium aluminide

Titanium-aluminide alloys based on TiAl have an excellent potential to become one of the most important aerospace materials because of their low density, high melting temperature, good elevated-temperature strength and modulus retention, high resistance to oxidation and hydrogen absorption, and excellent creep properties. The chief roadblock to their application is poor ductility at low to intermediate temperatures that results in low fracture toughness and a fast fatigue-crack growth rate. During the last several years, a great deal of effort has been made to improve these ductile properties. These endeavors have met with some success through chemistry modification and microstructure control.     

Direct production of alloys based on titanium aluminides

Direct production of alloys based on titanium aluminides has been demonstrated via two-step aluminothermic reduction of titanium tetrachloride (TiCl₄). In the first step, TiCl₄ is reduced by aluminium to titanium trichloride (TiCl₃) at a temperature below the boiling point of TiCl₄ and at 1 atm under argon. The resulting TiCl₃ intermediate product is then reacted with aluminium at temperatures up to around 1000 °C, leading to synthesis of a Ti–Al alloy powder. In addition, alloying additives such as Nb, Cr, Mo and C have also been included in various concentrations.     

Optimization of pulsed TIG cladding process of stellite alloy on carbon steel using RSM

Stellite 6 is a cobalt-base alloy which is resistant to wear and corrosion and retains these properties at high temperatures. The exceptional wear resistance of Stellite 6 is mainly due to the unique inherent characteristics of the hard carbides dispersed in a Co-Cr alloy matrix. In this study, pulsed tungsten inert gas (TIG) cladding process was carried out to deposit Stellite 6 on plain carbon steel plate. The beneficial effects of this cladding process are low heat input, low distortion, controlled weld bead volume, less hot cracking tendency, less absorption of gases by weld pool and better control of the fusion zone. The dilution effect is a key issue in the quality of cladded layers and, in this regard, the pulsed current tungsten inert gas (PCTIG) was performed to decrease excess heat input and melting of substrate. This paper deals with the investigation of the hardness and dilution ratio by experimental study and mathematical modeling. The experiments were conducted based on four-factor, five-level central composite rotatable design. The second-order regression method was developed to study the correlations. The developed models were checked for their adequacy and significance by ANOVA analysis and confirmation tests were carried out to check the accuracy of predicted values.     

基于支持向量机的焊缝尺寸预测

焊缝尺寸是决定焊接接头强度及有关性能的重要因素,因此也是焊接质量控制的重要内容.分别以焊接电流、电弧电压以及焊接速度作为输入向量构造样本集,建立CO2焊接焊缝尺寸支持向量机模型,分别运用线性核函数,多项式核函数、高斯径向基核函数以及指数径向基核函数对焊缝熔宽、焊缝熔深以及焊缝余高进行预测.结果表明,采用指数径向基核函数所建立的支持向量机模型可以有效地对焊缝尺寸进行预测,为进一步实现焊缝质量的在线控制提供依据.     

基于电流变智能材料的触觉显示器

设计了一种基于电流变液的触觉显示器,并对该触觉显示器的触摸敏感性进行了研究.该显示器具有5×5个显示单元,其表面是一层导电弹性薄膜,作为电极为电流变液提供电场.对该触觉显示器的表面触摸力响应实验研究表明,该触觉显示器显示能力持久且具有良好的重复使用性.通过监测通过电流变液层的电流变化,可以有效地探测出施加于触觉显示器表面的外力的变化.这个结果对进一步将该触觉显示器改造成兼具传感与作动功能于一体的新型阵列式触觉显示器具有极大的意义.