基于改进灰色神经网络模型的板料成形缺陷预测研究

为了准确预测和减少板料成形过程中可能出现的缺陷,提出了一种改进的灰色神经网络预测模型。该模型利用BP神经网络辅助灰色预测模型进行预测。其中,灰色模型进行粗预测,神经网络模型修正其误差,再通过寻找最佳权值以优化灰色模型中微分所对应的背景值,进而得到精度更高的灰色神经网络模型。以国际著名板料成形数值模拟会议NUMISHEET'93的方盒件拉深为例,运用改进的灰色神经网络模型,预测其拉裂和起皱。结果表明,改进的灰色神经网络模型具有很高的预测精度,相比于未改进的灰色神经网络模型,预测结果更加准确和稳定。     

Application of the hydroforming strain- and stress-limit diagrams to predict necking in metal bellows forming process

In order to predict the initiation of necking in metal bellows forming process, a methodology for determination of the forming limit diagram and the forming limit stress diagram is represented in this paper. The methodology is based on the Marciniak and Kuczynski (M–K) model. Comparison between the experimental and theoretical results for hydroforming stress and strain-limit diagrams as predicted by different methods indicates that the present approach is suitable for prediction of necking in tube hydroforming processes. Afterwards, the implementation of the hydroforming strain- and stress-limit diagrams into finite element numerical simulations for the forming of the metal bellows is established. A satisfactory agreement between the finite element method (FEM) and test results is achieved.     

Analysis and finite element simulation of the tube bulge hydroforming process

To investigate the effect of the loading path on the forming result and get the reasonable range of the loading path in tube bulge hydroforming process, a mathematical model considering the forming tube as an ellipsoidal surface is proposed to examine the plastic deformation behavior of a thin-walled tube during the tube bulge hydroforming process in an open die, and thus different loading paths are gained based on this model. The finite element code Ls-Dyna is also used for simulating the tube bulge hydroforming process. The effect of the loading paths on the bulged shape and the wall thickness distribution of the tube are discussed, and then the reasonable range of the loading path for the tube bulge hydroforming process is determined.     

Analysis and finite element simulation of the tube bulge hydroforming process

To investigate the effect of the loading path on the forming result and get the reasonable range of the loading path in tube bulge hydroforming process, a mathematical model considering the forming tube as an ellipsoidal surface is proposed to examine the plastic deformation behavior of a thin-walled tube during the tube bulge hydroforming process in an open die, and thus different loading paths are gained based on this model. The finite element code Ls-Dyna is also used for simulating the tube bulge hydroforming process. The effect of the loading paths on the bulged shape and the wall thickness distribution of the tube are discussed, and then the reasonable range of the loading path for the tube bulge hydroforming process is determined.     

Prediction of Deformed Configuration and Ductile Fracture for Simple Upsetting Using an Artificial Neural Network

This paper suggests a scheme for simultaneously accomplishing the prediction of fracture initiation and geometrical configuration of deformation in metal forming processes using an artificial neural network. A three-layer neural network is used and a back-propagation algorithm is adapted to train the network. The Cockcroft–Latham criterion is used to estimate whether fracture occurs during the deformation process. The geometrical configuration and the value of ductile fracture are measured by the finite-element method. The predictions of the neural network and the numerical results of simple upsetting are compared. The proposed scheme has predicted the geometrical configuration and fracture initiation successfully.     

弯曲轴线薄壁焊管内高压成形壁厚与变形规律的研究*

为了揭示焊缝对弯曲轴线类管件内高压成形的影响及缺陷产生的机制,采用试验和数值模拟的方法研究弯曲轴线焊管内高压成形的主要缺陷及壁厚分布规律,并分析焊缝在不同工序间的综合影响。结果表明,即使焊缝远离圆角区域,焊缝仍然是缺陷易发部位,弯曲使焊缝塑性下降,并导致在后续的工序中发生起皱甚至开裂。对于弯曲轴线薄壁焊管内高压成形,壁厚主要受弯曲和高压整形工序的影响,预成形工序对壁厚影响不大,而且焊缝的壁厚变化量始终小于其他区域。由此可知,焊缝是导致弯曲轴线薄壁焊管内高压成形缺陷产生的重要影响因素,将焊缝置于轻微压缩变形部位,是克服焊接接头性能下降导致的成形能力不足和避免缺陷产生的有效手段。     

Analytical model for planar tube hydroforming: Prediction of formed shape, corner fill, wall thinning, and forming pressure

An analytical model for planar tube hydroforming based on deformation theory has been developed. This analytical model can be used to predict hydroformed shape, corner fill, wall thinning, and forming pressure. As the model is based on a mechanistic approach with bending effects included, local strain and stress distribution across the wall thickness can be determined. This includes strain and stress distributions for the outer layer, inside layer, and middle layer. The model is validated using finite element analysis and tube hydroforming experiments on irregular triangular, irregular quadrilateral, and pentagonal hydroformed shapes.     

基于BP神经网络的TC4钛合金超塑性变形后组织及性能预测研究

分别用Visual Fortran语言和MATLAB软件建立了TC4钛合金超塑性变形时变形参数与其力学性能和晶粒尺寸之间的BP神经N络模型，通过用较少的力学性能和晶粒尺寸的试验数据进行训练，进而对其性能进行预测。结果表明，BP神经网络用于材料超塑性变形后的力学性能及晶粒尺寸预测是可行的，其预测误差小于7%。     

汽车后延臂液压胀形的数值模拟

在基于刚塑性模型的HydroFORM-3D有限元平台上,对汽车后延臂的液压胀形工艺进行数值模拟分析和研究,分别考察了汽车后延臂液压胀形的主要工艺参数,如管内压力、轴向进给量及摩擦条件对成形过程的影响。并与实际试验结果进行验证和比较,结果表明,数值模拟具有较好的计算精度,可以准确地预测出几何变形和厚度变薄量。     

Predicting axial piston pump performance using neural networks

A neural network model for an axial piston pump (bent-axis design) is derived in this paper. The model uses data obtained from an experimental setup. The purpose of this ongoing study is the reduction of the power loss at high pressures. However, at the beginning, a study is being done to predict the behavior of the current design of the pump. The neural network model has a feedforward architecture and uses the Levenberg–Marquardt optimization technique in the training process. The model was able to predict the behavior of the pump accurately.     

Collapse Pressure and Strength Analysis of Hydroformed Circular Plates

Hydroforming processing of sheet metal parts provides a number of advantages over conventional processes including the reduction in the number of parts, and reduction in the tooling and material costs. As a result, the technology has attracted increasing attention in many industries. However, there is still little experience available of both the process and its associated stresses and deformation which limit the application of this technique. The current study has been initiated to gain a better understanding of the fundamentals of the hydroforming process and the limitations on its applicability. The study outlines the stress and deformation distributions when a circular plate is processed by a hydrostatic transverse pressure that causes the plate to deform in the axial direction. The collapse load that leads to excessive and uncontrollable deformation is also determined and a forming limit is established. Experimental verification is also carried out to establish the validity of the analytical results.     

Bursting failure prediction in tube hydroforming processes by using rigid–plastic FEM combined with ductile fracture criterion

The most common failure in tube hydroforming is the bursting failure due to excessive thinning of large deformation. To evaluate the forming limit of hydroforming processes, the Oyane's ductile fracture integral I was introduced and calculated from the histories of stress and strain according to every element by using the rigid–plastic finite element method. The region of fracture initiation and the forming limit for three hydroforming processes, such as a tee extrusion, an automobile rear axle housing, and a lower arm under different forming conditions are predicted in this study. Also it is shown that the material parameters used in the ductile failure can be obtained from the experimental forming limit diagram. From the results, the prediction of the bursting failure and the plastic deformation for the three hydroforming examples demonstrates to be reasonable so that this approach can be extended to a wide range of practical tube hydroforming processes.     

Development of a hydroforming setup for deep drawing of square cups with variable blank holding force technique

Sheet hydroforming is a process that uses fluid pressure for deformation of a blank into a die cavity of desired shape. This process has high potential to manufacture complex auto body and other sheet metal parts. Successful production of parts using hydroforming mainly depends on design aspects of tooling as well as control of important process parameters such as closing force or blank holding force (BHF) and variation of fluid pressure with time. An experimental setup has been designed and developed for hydroforming of square cups from thin sheet materials. Square cups have been deep drawn using constant and variable BHF techniques. A methodology has been established to determine the variable BHF path for successful hydroforming of the cups with the assistance of programmable logic controller and data acquisition system. Finite element (FE) simulations have also been carried out to predict formability with both of these techniques. It has been found that it is possible to achieve better formability in terms of minimum corner radius and thinning in the case of variable BHF technique than in the case of constant BHF technique (constant force during forming and calibration). The results of FE analysis have been found to be in good agreement with experimental data.     

Experimental and numerical study of stamp hydroforming of sheet metals

The objectives of this research was to experimentally and numerically study the stamp hydroforming process as a means for shaping aluminum alloy sheets. In stamp hydroforming, one or both surfaces of the sheet metal are supported with a pressurized viscous fluid to assist with the stamping of the part thereby eliminating the need for a female die. The pressurized fluid serves several purposes: (1) supports the sheet metal from the start to the end of the forming process, thus yielding a better formed part, (2) delays the onset of material failure and (3) reduces wrinkle formation. This paper focuses on the experimental and numerical results of the stamp hydroforming process utilizing a fluid pressure applied to one surface of the sheet metal. The effects of applying a constant, varying and localized pressure to the surface of 3003-H14-aluminum sheet alloy were evaluated. Experiments demonstrated draw depths improvements up to 31% before the material failed. A failure prediction analysis by Hsu was also carried out to predict an optimal fluid pressure path for the varying fluid pressure case. The commercial finite element analysis code Ls-Dyna3D was used to numerically simulate the stamp hydroforming process. Both isotropic and anisotropic material models were used and their predictions compared against the experimental results. The numerical simulations utilizing Barlat's anisotropic yield function accurately predicted the location of the material failure and the wrinkling characteristics of the aluminum sheet.     

瑞风商务车托架拉延成形数值模拟及工艺参数优化

基于逆向工程建立瑞风商务车托架零件的几何模型,并基于Dynaform软件平台对不同工艺参数下该零件的拉延成形过程进行数值模拟。在此基础上,以压边力、拉延筋高度和拉延筋圆角半径作为设计变量,以零件不发生破裂为优化目标,以有限元数值模拟结果作为虚拟样本,建立目标函数的人工神经网络预测模型;将人工神经网络预测模型作为优化算法的知识源,采用遗传算法对压边力、拉深筋几何参数等工艺参数进行了优化设计。试验结果表明,数值模拟、神经网络预测和工艺优化是可靠的,从而可为制定金属板料最佳的冲压成形工艺提供一条先进、合理的途径。     

内高压成形机理与关键技术

在中国国家杰出青年科学基金资助项目“镁合金热态液力成形技术”、中国国家自然科学基金资助项目“轻体件高内压液力成形机理的研究”、“管材热态内压成形新方法及其机理研究”和“激光拼焊管内高压成形机理”、以及中国教育部高等学校博士学科点专项科研基金资助项目“镁合金热态内高压成形机理研究”共同资助下,开展内高压成形机理及关键技术研究,在内高压成形塑性变形规律、起皱和破裂等失稳行为、提高成形极限和降低成形压力方法,以及液力胀接、热态内压成形和拼焊管内高压成形等方面取得重要进展,并在汽车和航天等领域实现内高压成形技术产业化应用,报告上述研究的理论和工程体系。 根据塑性变形特点,将内高压成形分为变径管内高压成形(IHPF of TPVD)、弯曲轴线管内高压成形(IHPF of TPCA)和多通管内高压成形(IHPF of TPB/BT)等3类,提出IHPF of TPVD由充填、成形、整形等步骤组成,IHPF of TPCA由弯曲、预成形、内高压成形等步骤组成,IHPF of TPB/BT由胀形、补料、整形等步骤组成。以此为出发点,通过实验和理论分析,研究IHPF塑性变形规律与失稳行为。     

Prediction of Rolling Force and Deformation in Three-Dimensional Cold Rolling by Using the Finite-Element Method and a Neural Network

In this study of a 3D rolling process, an incremental updated Lagrangian elasto-plastic finite-element method and a neural network have been integrated to predict the rolling force and the maximum surface error during the rolling process. A series of 27 sets of tools of different geometry were used for simulation of the rolling process, with different variations of die radius, rolling ratio, and matching for the different thicknesses of the products. The results of the rolling force and of the deformation of the surface are then input to a neural network to establish a model for the rolling variables. The results of rolling processing by this developed abductive network can be accurately predicted, once the rolling control parameters are given. This work achieved a satisfactory result based on a demonstration of the simulation, proving that this is a new and feasible approach which can be used for control of the rolling process for materials. ID="A1"Correspondance and offprint requests to: Professor J. C. Lin, Depart-mentof Mechanical Design Engineering, National Huwei Institute of Technology, Yunlin 632, Taiwan. E-mail: linrc@sunws.nhit.edu.tw     

Modelling pressure distribution in a rectangular gas bearing using neural networks

Gas ubricated bearings are of tremendous use especially in the biomedical and aerospace industries. For that reason, gas bearings have been the subject of much research for the past decade or so. Experimental as well as theoretical work has been done to calculate the pressure distribution inside the bearing. The models available to predict the pressure are primitive and need to be improved. This paper discusses a new modelling scheme known as artificial neural networks. The pressure distribution and the load-carrying capacity are predicted using feedforward architecture of neurons. The inputs to the networks are a collection of experimental data. This data is used to train the network using the Levenberg-Marquardt optimization technique. The results of the neural network model are compared to a theoretical model and the results are promising. The neural network model outperforms the avallable theoretical model in predicting the pressure as well as the load-carrying capacity.     

考虑接触演变的板材回弹过程建模与优化

回弹是由工件在卸载后的弹性变形引起的。板料成形过程中为了控制成形件的最终形状,必须进行回弹设计优化。准确预测回弹对于板料成形过程的模具设计非常重要。降低回弹模拟结果与试验结果的偏差是设计过程中的难题。基于NUMISHEET’02的自由弯曲标准考题考虑板材与模具间的接触演变过程,建立了一个有限元模型来预测回弹。采用一个常规的优化方法对有限元分析中的材料和单元模型进行了分析,研究发现不同模型对回弹结果有较大影响。模拟结果与参考文献中的试验结果比较表明了模型的正确性和可行性。     

薄壁半球件无模液压胀形的实验研究

对薄壁半球形件的无模液压胀形工艺进行了研究，并对壳体成形过程中的变形规律进行了分析，实验证明，采用无模液压胀形工艺形半球件是完全可行的，为大型薄壁半球形零件的加工制作提供了一种全新的成形方法。