铝型材挤压模工作带优化

基于MATLAB平台，将BP神经网络、遗传算法和数值模拟技术应用于铝型材挤压模具参数优化设计。采用三层BP神经网络建立型材挤压模具的数学模型，由正交实验法安排模拟实验组合，采用有限元软件进行挤压过程的数值模拟，并以具有不同工作带尺寸的挤压模具中金属流出模口平面上的Z向质点流速均方差作为模型目标值，将模拟结果作为神经网络的输人样本对训练网络并建立网络知识源，通过遗传算法求得模型的全局优化解；最后通过有限元数值模拟技术验证并比较优化所得工作带与经验法确定的工作带对金属流动均匀性的影响。数值模拟结果表明，本研究对挤压模具工作带的优化是有效的。     

异形铝型材挤压模具受力分析与工艺优化

以某空心异形铝型材为研究对象,设计了挤压分流模。基于Hyperxtrude软件对挤压过程进行模拟,得到了铝型材挤压模具变形分布和等效应力分布。基于模拟结果对挤压工艺参数进行了正交试验设计和优化。结果表明,模具的等效应力降低了。结果可对异形空心铝型材挤压模具设计和工艺参数选择提供参考。     

轨道车辆用超宽幅薄壁铝型材挤压过程数值模拟及模具优化

以轨道车辆用某扁宽、薄壁、弧形的大型铝合金型材为研究对象,基于HyperXtrude专业铝型材挤压成型有限元分析平台,采用拉格朗日-欧拉(ALE)算法,对其稳态挤压成型过程进行数值模拟。结果表明,采用初始模具挤压时,型材出口处的金属流速不均匀。通过增设阻流坎和扩大供料容积,多次优化模具结构后,型材出口处金属流速均匀。数值模拟结果与生产试模结果吻合,数值模拟结果可以为大型复杂截面铝型材挤压模具结构设计和优化提供指导。     

铝型材出口速度分布的数值模拟

铝型材挤压过程中的出口速度,是模具优化设计的重要参数之一,型材出口速度不均的原因之一是模具设计不当。为了获得高质量的挤压制品,对模具出口区域速度分布的研究是必须的。本文采用DEFORM3D对铝型材进行了3D模拟,获得了在模具制造之间铝型材的出口速度。数值模拟的结果得到了实验的验证,这对优化模具过程是非常重要的。     

铝型材悬臂类模具挤压成形数值分析与研究

基于HyperXtrude铝型材模具热挤压仿真系统,以简单U形悬臂模为例,对比研究了不带导流和带导流两种不同铝型材悬臂类挤压模具结构在稳态挤压过程中的型材流速、型材质量、模具强度等方面所存在的普遍规律,获得了与实际一致的结果。     

平板铝型材宽展挤压模结构改进

针对某企业生产的平板铝型材氧化后有夹渣、组织条痕等缺陷,采用弹塑性有限元理论,对铝型材宽展挤压过程进行了数值模拟。根据模拟结果分析了造成该缺陷的主要原因,并对生产平板铝型材的宽展挤压模的结构进行改进和优化。通过生产验证,挤压出质量合格的平板铝型材,提出的模具改进方案是行之有效的。     

铝型材宽展挤压模具参数优化模型

基于MATLAB平台,提出了一种集数值模拟、BP神经网络和遗传算法为一体优化模型,用于型材宽展挤压模具结构参数优化.以FR029铝型材的宽展挤压为例,合理选择宽展模具入口宽度和出口宽度尺寸.通过有限元模拟验证,说明提出的工艺参数优化模型是行之有效和正确的,对指导型材宽展挤压模具优化设计具有重要意义.     

基于HyperXtrude的轨道交通铝型材挤压模具仿真模拟

以轨道车辆用某一扁宽、薄壁、空心、弧形大型铝合金型材为研究对象,根据该型材特点制定模具设计方案,采用基于ALE算法的Hyper Xtrude挤压模拟分析软件对模具挤压过程进行模拟,将数值模拟结果与实际挤压结果进行对比,发现数值模拟结果与生产试模结果吻合。得出的数值模拟结果可以为大型复杂截面铝型材挤压模具结构设计和优化提供指导。     

铝合金软管挤压模拟及模具优化方法研究

铝型材挤压模具是保证铝产品质量的重要工具。为了合理地设计出挤压模具结构,利用hyperxtrude软件对挤压过程进行了模拟。根据模拟挤压过程中温度场、速度场、应力应变场等,验证了传统经验设计的模具,并提出了优化方案。这大大缩短了模具设计过程,能为企业提高生产效率提供实用参考。     

铝型材挤压成型数值模拟与模具优化设计

采用HyperXtrude专业铝型材挤压成型有限元软件系统,以多边型空心铝型材为研究对象,对其稳态挤压成型过程进行数值模拟,计算结果发现初始模具结构中金属材料流速严重不均匀.通过增设导流槽、合理布置分流孔和调整工作带长度等措施,获得了较为理想的模具结构,同时提高了模具的寿命.     

Application of neural network and FEM for metal forming processes

This paper proposes a new technique to apply the artificial neural network in metal forming processes. A three-layer neural network is used and a back propagation algorithm is employed to train the network. It is determined by applying the ability of function approximation of the neural network to the initial billets which satisfy the minimum of incomplete filling in the die cavity. The die geometry for cylindrical pulley is designed to satisfy the design conditions of the final product. The proposed schemes have been successfully adapted to find the initial billet size for axisymmetric rib-web product and to design the die geometry for cylindrical pulley. The neural network may reduce the number of finite element simulation for designing the die of forging products and further it is usefully applied to multi-stage process planning.     

基于BP人工神经网络的油箱端盖拉深成形仿真预测

以油箱端盖作为分析对象,借助DYNAFORM仿真软件,对油箱端盖的拉深成形过程进行数值模拟,并通过拉深成形试验验证可知,板料最大减薄率与最大增厚率的试验值与模拟值之间的相对误差分别为9.26%与8.32%,验证了有限元模型的正确性。结合正交试验,进行有限元仿真试验的设计,基于BP人工神经网络,对板料的成形质量进行仿真预测。选择冲压速度、模具间隙以及压边力作为输入层,将板料成形的最大减薄率作为输出层,建立了3-11-1的3层BP人工神经网络。通过BP人工神经网络的训练与测试得知:BP人工神经网络仿真预测值与数值模拟值之间的相对误差为2.15%,验证了BP人工神经网络应用于油箱端盖拉深成形质量仿真预测的正确性。     

Void closure prediction in cold rolling using finite element analysis and neural network

Cold rolling is used to eliminate void defects in cast materials thus improving the material performance during service. A comprehensive procedure is developed using finite element analysis and neural network to predict the degree of void closure. A three-dimensional nonlinear dynamic finite element model was used to study the mechanism of void deformation. Experiments were conducted to investigate void closure during the cold flat rolling process. Experimental results are compared to the three-dimensional finite element predictions to validate the model. The void reduction predictions from finite element analysis are in good agreement with experimental findings. Plastic strain, principal stress distribution around the void and void reduction ratio are presented for various case studies. As finite element simulation is time-consuming, a back-propagation neural network model is also developed to predict void closure behavior. Based on the correlation analysis, the reduction in sheet thickness, the dimension of the void and the size of the rollers were selected as the inputs for the neural network. The neural network model was trained based on results obtained from finite element analysis for various simulation cases. The trained neural network model provides an accurate and efficient procedure to predict void closure behavior in cold rolling.     

交互式挤压模具智能优化设计系统研究

在高效无敏度智能优化算法及挤压成型过程数值模拟的基础上,结合刚—黏塑性有限元模拟、神经网络、遗传算法,提出一种交互式挤压模具智能优化设计方法。并利用面向对象编程技术C++语言,在VC++6.0程序开发器上开发出了界面友好、自动化程度高、模块通用性好的交互式挤压模具优化设计系统,实现了挤压过程中模具设计多种目标的优化控制,并详细阐述了系统的结构功能、关键技术及应用实例。由于数值模拟、人工神经网络和遗传算法属于非耦合算法,通用性较强,该系统适用于金属塑性成形各种领域的优化设计。     

基于遗传神经网络的压铸机座板结构优化设计

在对压铸机合模机构进行结构设计时，利用神经网络的非线性映射能力，通过少量样本的有限元分析结果，训练出表述结构参数间函数关系的神经网络模型，然后利用遗传算法的全局寻优性找到神经网络模型表述的目标函数的最优结构参数，从而解决结构优化设计的瓶颈和智能问题，利用这种优化设计策略，设计了压铸机合模机构座板，结果表明了该方法的高效性。     

The application of neural networks in the preform design of the upsetting process

Design of the optimum preform for near net shape manufacturing is a crucial step in upsetting process design. In this study, artificial neural networks (ANN) are used to consider different interfacial friction conditions between the top and bottom die and billet interface. Two back propagation neural networks are trained based on finite element analysis results considering ten interfacial friction conditions and varying geometrical and processing parameters, to predict the optimum preform for high strength (HS) steel and commercial aluminum. Neural network predictions were verified for three new problems of both HS steel and commercial aluminum and observed that these are in close match with their simulation counterparts. It was further Experimentally verified with two commercial aluminum specimens and observed that the preform values predicted by ANN are in good agreement with experimental results.     

薄壁变形机匣件限位检测机理分析与预测

为了研究带有一定公差的薄壁机匣件二次装夹变形规律,以约束表面和相关表面为研究对象,建立了机匣件约束表面与相关表面变形关系的数学分析模型。使用有限元法模拟变形机匣件的限位装夹过程,将模拟产生的数据作为BP神经网络的训练样本,借助神经网络的非线性映射能力建立了装夹变形的预测模型。通过对某航空材料7075机匣件作试验分析,验证了所构建的三维有限元模型与神经网络预测模型较为合理,能够为机匣件的加工制造提供反馈,为机匣件的装配分析提供良好的校验。     

Application of artificial neural network and Taguchi method to preform design in metal forming considering workability

This study describes a new method of perform design in muti-stage metal forming processes considering workability limited by ductile fracture. The finite element simulation combined with ductile fracture criterion has been performed in order to predict ductile fracture. The artificial neural network using the Taguchi method has been implemented for minimizing objective functions relevant to the forming process. The combinations of design parameters used in finite element simulation are selected by orthogonal array in statistical design of experiments. The orthogonal array and the result of simulation are used as train data for artificial neural networks. The cold heading process is taken as an example of designing preforms which do not form any fracture in the finished product. The results of analysis to validate the proposed design method are presented.     

TC4钛合金神经网络本构模型及在有限元模拟中应用

利用Zwick/Roell Z100材料试验机,对TC4钛合金进行等温恒应变速率下的单向拉伸试验。基于获得的试验数据,采用BP神经网络技术建立了该合金的高温本构关系模型,并对其预测性能进行分析。基于ABAQUS/Explcit平台进行材料子程序二次开发,将神经网络本构模型嵌入到有限元计算中,实现了TC4钛合金高温变形的数值模拟。结果表明,神经网络本构模型预测精度很高,可以准确地描述TC4钛合金在热态下的动态力学性能。神经网络本构模型应用于有限元模拟可行且有效。