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.     

阀用位移传感器的多目标优化设计与仿真

为解决阀用位移传感器多目标多准则的电磁结构耦合优化设计问题,针对径向尺寸受到严格约束的长行程、高精度阀用位移传感器,运用MATLAB编写优化算法程序,得到以线性度和灵敏度为目标、以结构尺寸参数为约束条件的最优化参数匹配组。根据多目标优化结果,运用Maxwell电磁仿真软件对阀用LVDT进行等效建模,对优化模型和原始模型进行电磁场仿真分析,对比两者的输出特性。结果表明：优化模型的线性度为0.19%,灵敏度为23.12 mV/mm,较原始模型有较大的提高。该方法较完整地解决了LVDT结构优化设计,可为小型化阀用位移传感器的优化与设计提供参考。     

金属体积成形预成形设计的现状及发展

金属体积成形预成形设计 ,是材料多工序成形工艺与模具设计的中心工作。如何设计预锻件形状和预锻模形状是生产合格终锻件的关键 ,也是模具设计的难点之一。预成形优化设计是采用了优化技术的预成形设计 ,是发展的预成形设计 ,在国内外得到了广泛研究。本文首先介绍了预成形设计所依据的知识和采用的主要手段 :基于以往丰富经验的经验性指导原则、物理模拟技术和基于数值模拟的计算机辅助设计。然后对处于当今预成形设计的主流方法—基于数值模拟的预成形设计的两个发展阶段 ,即预成形设计和预成形优化设计进行了介绍。其次对常用的正向模拟和反向模拟预成形设计方法和常用的基于灵敏度分析的正向模拟优化法、直接微分法、拟合优化法、微观遗传算法等预成形优化设计方法的基本原理、特点及应用进行了介绍和总结。最后指出了预成形设计的发展趋势。     

基于实验设计的电火花微小孔加工参数优化研究

综合考虑电火花微小孔加工中伺服参数与电参数的影响，采用实验设计的方法对微小孔加工效率进行优化研究。首先以一次正交试验对小孔加工中的微进给速度进行优化，以获得稳定快速的伺服响应，在此基础上进行二次正交试验，采用田口方法确定电参数的最佳水平组合。经参数优化后，小孔加工时间由初始的137．2s缩短到67．9s，加工效率显著提高。     

Sensitivity analysis based preform die shape design using the finite element method

This paper uses a finite element-based sensitivity analysis method to design the preform die shape for metal forming processes. The sensitivity analysis was developed using the rigid visco-plastic finite element method. The preform die shapes are represented by cubic B-spline curves. The control points or coefficients of the B-spline are used as the design variables. The optimization problem is to minimize the difference between the realized and the desired final forging shapes. The sensitivity analysis includes the sensitivities of the objective function, nodal coordinates, and nodal velocities with respect to the design variables. The remeshing procedure and the interpolation/transfer of the history/dependent parameters are considered. An adjustment of the volume loss resulting from the finite element analysis is used to make the workpiece volume consistent in each optimization iteration and improve the optimization convergence. In addition, a technique for dealing with fold-over defects during the forming simulation is employed in order to continue the optimization procedures of the preform die shape design. The method developed in this paper is used to design the preform die shape for both plane strain and axisymmetric deformations with shaped cavities. The analysis shows that satisfactory final forging shapes are obtained using the optimized preform die shapes.     

Quantification of micro-cracks on the bending surface of roll formed products using the GTN model

With the rising interest in lightweight construction, the usage of high strength steel has increased remarkably during the last years. Unfortunately, there is manufacturing problem: high strength steel has a higher occurrence rate of cracks than mild steel. The Gurson-Tvergaard-Needleman (GTN) model is used to precisely predict the occurrence of cracks in roll forming. The flow stress and the GTN material parameters are identified using a load-displacement curve obtained from tensile test. In detail, the material parameters that have to be determined were selected and based on this 3-level orthogonal array is made. The coefficients of the objective function are derived using the response surface method with the orthogonal array based on the simulation results. The material parameters are obtained by minimizing the objective function and determined by comparing the simulation and test results. The reliability of the material parameters determined by the tensile test increased through cross-validation using the bending test. The roll forming simulation with the determined parameters is fulfilled, and the experiments are performed to determine the occurrence of micro-cracks according to bending angles. Micro-cracks are quantified by comparing the void volume fraction (f) from the GTN model, and the micro-cracks on the surface are measured using a scanning-electron-microscope.     

A probabilistic approach for the optimisation of ultrasonic array inspection techniques

Ultrasonic arrays are now used routinely for the inspection of engineering structures in order to maintain their integrity and assess their performance. Such inspections are usually optimised manually using empirical measurements and parametric studies which are laborious, time-consuming, and may not result in an optimal approach. In this paper, a general framework for the optimisation of ultrasonic array inspection techniques in NDE is presented. Defect detection rate is set as the main inspection objective and used to assess the performance of the optimisation framework. Statistical modelling of the inspection is used to form the optimisation problem and incorporate inspection uncertainty such as crack type and location, material properties and geometry, etc. A genetic algorithm is used to solve the global optimisation problem. As a demonstration, the optimisation framework is used with two objective functions based on array signal amplitude and signal-to-noise ratio (SNR). The optimal use of plane B-scan and total focusing method imaging algorithms is also investigated. The performance of the optimisation scheme is explored in simulation and then validated experimentally. It has been found that, for the inspection scenarios considered, TFM provides better detectability in a statistical sense than plane B-scan imaging in scenarios where uncertainty in the inspection is expected.     

超声相控阵声束控制特性分析

超声相控阵技术通过对超声阵列换能器中各阵元进行相位控制,获得灵活可控的合成波束,从而实现对整个检测对象的扫描。它具有可进行动态聚焦、可成像检测、可检测复杂形状物体等一些优点,近年来正成为国际无损检测领域的研究热点,并已经在医学和工业领域得到应用。在研究超声相控阵技术原理的基础上,建立了数字声束模型来分析线阵换能器参数对声束控制的影响,并用计算机软件进行仿真分析声束的偏转聚焦。通过这种方法,可以很清楚的了解超声相控阵声束控制特性。     

挤压锻造模具形状的混合智能优化

数值仿真技术已成为挤压锻造模具设计中的重要评价手段，有助于设计人员更好的理解设计空间。然而，由于计算时问长，目标函数无法显式表述，搜索空间可能不连续等因素，很难直接基于数值仿真通过梯度法进行设计优化。非梯度的优化方法比如遗传算法、模拟退火法等由于需要更多的仿真次数，也很难实现。本文基于人工智能中的归纳学习方法，提出了一种针对挤压锻造工艺优化的基于数值仿真归纳知识的混合优化方法，并通过一个挤压锻造的实例验证了该方法的有效性。     

VHT800立式车铣加工中心立柱结构静动态优化及轻量化设计

研究了VHT800立式车铣加工中心立柱结构的静动态优化及轻量化设计方法,解决了企业经验设计结构存在的静、动态性能不足以及结构过于笨重的问题.首先基于机床实际典型工况下的立柱载荷及边界条件,运用拓扑优化技术建立了体积约束下的立柱结构多工况静态性能和动态性能多目标拓扑优化概念模型;然后对概念模型进行结构刚度设计及工艺设计.最终为企业提供了新的立柱结构方案,该方案的结构刚度和一阶固有频率比经验设计结构分别提高了13.6%和17.1%,同时质量减轻了9.7%,表明该设计方法能够有效提高机床立柱结构静、动态性能并实现结构轻量化设计.     

基于软测量的可重构模块机器人分散力/位置控制

在受限空间内,可重构模块机器人末端腕力传感器造价昂贵且易受环境影响,它的使用还提高了机器人在软硬件设计上的复杂程度,针对这一问题,提出了基于软测量的可重构模块机器人分散力/位置控制方法。将可重构模块机器人系统在各个方向上进行模型分解,各方向的模型看作一个子系统,利用自适应RBF神经网络去估计各子系统间的不确定项和耦合关联项。在末端未安装腕力传感器的情况下,利用自适应RBF神经网络去估计末端接触力,并推导出了RBF神经网络的权值、径向基函数中心和宽度的自适应律。2类不同自由度可重构模块机器人的仿真结果验证了所提方法的有效性。     

基于混合粒子群算法的圆柱型电磁铁优化设计

为了解决当电磁铁外型相同如何使电磁力最大的问题,将现有电磁铁设计思路应用到圆柱型电磁铁设计中,并采用改进的混合粒子群方法优化电磁铁内部结构参数。运用等效磁路法推导出电磁铁数学模型,并通过Maxwell有限元仿真对比验证准确性,然后采用单目标优化方法,得到影响电磁力的参数,最后采用改进的粒子群算法进行多目标优化,得到优化后的参数值。得到的几组优化数据表明：多目标优化方法得到的有效工作区域电磁力更大,仿真结果表明优化后的电磁力综合提高20%,证明优化方法有效。     

3D封装芯片焊点可靠性有限元分析

选择工业中广泛使用的Sn,Sn3.9Ag0.6Cu钎料作为三维封装芯片键合焊点材料,采用ANSYS有限元软件建立三维封装模型,基于Garofalo-Arrhenius本构方程,进行-55～125°C热循环模拟,并通过田口法探讨封装结构和工艺参数对焊点的可靠性影响.结果表明,Cu柱与焊点接触处是整个结构的薄弱区域,在IM...     

Design of decentralized robust stabilizing controller for interconnected time-varying uncertain systems

lINTR0DUCTI0NInrecentyears,decentralizedcontrolforin-terconnectedlargesystemshasbeenpaidmuch.tt..ti..[lj.Sincethereexistsuncertaintiesinthesystemmodels,performancecannotbeat-tainedifthecontrollerdesignisonlybasedonnominalplant,whichresultsinthewidelystudyofrobuststabilizationfOruncertaininterconnect-edlargesystems['~'].However,inmostpapers,onlysufficientconditionsweregiveninRiccatie-qualitiesorinequalities,whichneededpreadjust-mentofparametersinadvance,andwasverycomplicatedandinconvenient.…     

Development and Performance of New Gadolinium and Boron Containing Radiation-Absorbing Composite Systems

The theoretical design and the experimental design and development of multifunctional radiation-absorbing composite material systems based on gadolinium, boron, and tungsten has been carried out. Based on theoretical calculations, the effective compositions of these composite subsystems were established for the enhanced absorption of neutron and γ irradiation in various energy spectra. In addition, the systems and systems compositions were designed and processed for enhanced multifunctional performance. Selected and optimized compositions of Gd-B-W system were densified by shock wave consolidation technology. The technological parameters for the explosive consolidation processes and the structure-properties relationships are presented and discussed. The radiation-absorbing properties of the bulk samples were investigated and measured under neutron and gamma irradiation. The theoretical design and the optimization of these composite systems were carried out by the Monte Carlo method with a GEANT 3 program, which contained a special GCALOR package for the simulation of the interaction of thermal and fast neutrons with the composite material systems. During the neutron passage through the samples, the main processes of the interaction of neutrons with matter were considered including elastic and inelastic scattering, neutron fission of nuclei, and radiation capture. The attenuation factor of the irradiation flux is determined as a criterion of efficiency of radiation absorption. For the energy of 0.025 eV (thermal neutrons), gadolinium-containing composites have the maximum absorption capability. In the range of energy spectrum from 1 eV to 10 eV, the boron-containing composites have better absorption performance. For the capture of neutrons in wide energy spectrum, the (n, γ) reaction takes place and tungsten provides enhanced absorption of radiation. In the presence of mixed radiation sources (neutron and γ quanta), the boron- and gadolinium-containing composite materials prepared on the tungsten basis have the best performance. In addition to enhanced radiation absorption properties, these composite systems show also enhancement on other properties such as corrosion resistance in aggressive media.     

Multi-objective optimization of process parameters during low-pressure die casting of AZ91D magnesium alloy wheel castings

Multi-objective optimization has been increasingly applied in engineering where optimal decisions need to be made in the presence of trade-offs between two or more objectives. Minimizing the volume of shrinkage porosity, while reducing the secondary dendritic arm spacing of a wheel casting during low-pressure die casting(LPDC) process, was taken as an example of such problem. A commercial simulation software Pro CASTTM was applied to simulate the filling and solidification processes. Additionally, a program for integrating the optimization algorithm with numerical simulation was developed based on SiPESC. By setting pouring temperature and filling pressure as design variables, shrinkage porosity and secondary dendritic arm spacing as objective variables, the multi-objective optimization of minimum volume of shrinkage porosity and secondary dendritic arm spacing was achieved. The optimal combination of AZ91 D wheel casting was: pouring temperature 689 °C and filling pressure 6.5 kPa. The predicted values decreased from 4.1% to 2.1% for shrinkage porosity, and 88.5 μm to 81.2 μm for the secondary dendritic arm spacing. The optimal results proved the feasibility of the developed program in multi-objective optimization.     

A combined model for the description of austenitization,homogenization and grain growth in hypoeutectoid Fe–C steels during heating

A combined model which allows one to simulate all the steps of the reaustenitization process of ferrito-pearlitic plain carbon steel has been developed. The dissolution of pearlite, the transformation of ferrite into austenite and the homogenization of the carbon distribution is described with a finite volume method. The simulation is performed on a bidimensional domain where ferrite (α), pearlite (P) and austenite (γ) grains are represented. The dissolution of pearlite is described by the growth of spherical grains and simple nucleation and growth laws. The movement of α/γ interfaces is calculated by solving the diffusion equation for carbon in the α and γ phases and accounting for the solute flux balance at the interface using a pseudo-front tracking method. The diffusion model is coupled with a Monte Carlo simulation which describes the grain growth occurring in austenite at a later stage of austenitization. The evolution of the volume fractions of pearlite and ferrite, the maximum and minimum carbon concentrations in the domain and the mean austenite grain size are represented as a function of the temperature for a typical case of constant heating rate. The influence of the different steps of the austenitization process on the global kinetics is discussed.     

基于近似替代模型的多工位高速锻造预锻模具优化

多工位高速锻造是一种节能减耗的先进成形技术,与传统的单工位多工序锻造有较大区别,不合适的预锻工位模具设计在生产中会造成载荷过大、锻件充填质量差、折叠等问题。文章提出了基于近似替代模型的多工位高速锻造预锻模具优化设计方法,采用拉丁超立方抽样方法构造初始样本点集,应用有限元软件获得响应值,并应用Kriging模型和BP神经网络模型建立真实问题的替代模型,来近似拟合预锻工位模具形状与成形载荷、成形质量之间的关系,并结合基于惩罚函数法的遗传算法求得最优解。以3工位高速锻造问题为例,经过优化取得了良好的效果,与初始设计相比,成形质量显著提高,成形载荷下降了40%左右。     

基于遗传算法的箱型结构焊接顺序优化

基于遗传算法原理,建立了箱型结构热-机耦合非线性三维优化仿真模型,以焊接变形为目标函数,进行了焊接顺序数值仿真优化.通过仿真计算优化结果与试验数据比较,证明所提出的优化方法与热-机模型结合是可行的,对于给定的焊接条件,通过选择合适的目标函数,采用遗传算法可以确定最优焊接顺序.这为针对工程具体问题设计焊接工艺提供了可靠的依据.     

Multi-objective optimization in material design and selection

The development or selection of a material to meet given design requirements generally requires that a compromise be struck between several, usually conflicting, objectives. The ways in which multi-objective optimization methods can be adapted to address this problem are explored. It is found that trade-off surfaces give a way of visualizing the alternative compromises, and that value functions (or “utility” functions) identify the part of the surface on which optimal solutions lie. The method is illustrated with examples.