一维下料问题的一种混合启发式算法

讨论一维下料问题,即用库存线材原料切割出若干种已知数量和长度的零件,优化目标为最大化线材利用率。提出一种混合启发式下料算法,用当前零件构造线材所有可能的切割方式,计算每种切割方式的废料长度、零件平均长度、大零件数量,按照废料最小、零件平均长度最大、大零件数量最多的原则选择一种切割方式切割相应线材,满足部分零件需求;用剩余零件继续构造线材可能的切割方式,按上述原则继续选择相应线材进行切割,重复该过程,直到所有零件的需求量得到满足为止。采用文献中基准例题验证文中下料算法,数值实验结果表明该算法与4种文献算法相比性能占优,且算法计算时间合理。     

一维下料问题的启发式方法

针对单一规格的一维下料问题,提出一种新的随机启发式搜索算法,在求出所有可能切割方式的情况下,搜索出最优解,而且保证最后一根原材料余料长度最长,方便以后下料.对于多规格原材料情况的一维下料分解为多个单一原材料的一维下料问题来进行求解.计算表明,与启发式算法或者遗传算法相比较,随机启发式搜索算法结构简明,易于编程,计算速度快,节材效果优.     

面向可加工性的矩形件优化下料方法

针对多规格、大批量的矩形件优化下料问题,综合考虑原材料利用率高、切割加工路径短等优化目标,提出一种面向可加工性的矩形件优化下料方法。运用矩形原材料单方向余料最小化计算方法优选出候选板条,候选板条构造过程中的共边排样可用于共边切割,缩短了矩形原材料上局部切割加工路径;利用基于候选板条的连续启发式排样算法实现矩形原材料上优化排样设计,保证高原材料利用率的同时,简化了矩形原材料上切割加工总路径。通过实例验证了所提方法的可行性和有效性。     

基于在线检测动态一维下料问题的GPU并行蚁群算法

随着在线检测技术发展,生产线上的物料需要根据检测结果进行快速切割。已有一维下料优化问题是根据全局目标进行建模的,其最优化算法不能满足实时调整切割方案的要求。本文首先根据物料在线检测及切割特点提出了动态多规格一维下料优化问题,并给出最优化模型;然后结合GPU特点创建并行蚁群算法来求解多规格动态一维下料问题,以保证在有限时间内求得近似最优结果;经过算法分析证明,对于大规模数据变量,并行蚁群算法效率高于传统蚁群算法。通过实验表明,在大规模数据量下,此并行蚁群算法与传统蚁群算法和分支定界算法相比,能够在较短时间内得到较优切割方案。     

递推矩阵的列生成算法下的一维下料方案研究

针对工程实际中一维下料优化问题,这里采用以原材料使用量最少为目标,以每根原材料的余料长度小于最短零件长度等为约束,建立数学优化模型,提出使用一种基于递推矩阵的列生成算法进行求解该模型.此算法主要通过求解列生成优化模型中的递推矩阵,生成添加列,最终得出下料方案.最后针对单一订单的情况,并根据实例优化计算所得结果进行对比,...     

蚁群算法在一维下料优化问题中的应用

研究了一维下料优化问题,建立了一维下料的双目标多约束数学优化模型,并给出了求解模型的蚁群算法的具体实现过程;针对模型的特殊性,提出了零件编码和将所有下料方式首尾连接转化为蚂蚁路径的方法,并讨论了初始化参数变量的选择问题。通过一维下料实例进行验证,计算结果表明,此算法可逼近理论最优解,收敛速度较快,理论上可以解决多规格大规模的一维下料优化问题。     

一维下料问题的多叉树遍历算法研究

切割条形材料,获得不同长度工件问题普遍存在。为满足原材料利用率最大的实际加工要求,将数据建立为多叉树模型,添加剪枝优化条件,使用深度优先遍历,递归调用深度优先遍历函数,搜索全局最优解,即最优化的下料方案。使用其他论文中的实例,对比启发式多级序列线性优化方法的运算结果。对比结果表明:多叉树遍历算法的材料利用率最高、结果更优。     

多目标一维下料决策方法研究

研究目标为最小化原料浪费、最小化下料方式数和最小化可用余料返回的多目标优化下料问题。运用多目标优化和多属性决策相结合的方法设计下料决策方法,即先用改进的非支配排序启发式进化算法求出问题的Pareto最优解集,再采用综合主客观赋权法计算各优化目标的权重,最后运用多属性决策方法选出一个满意解作为下料方案。实验结果证实所提方法对多目标下料决策是有效的。     

利用计算机进行优化下料

生产中常要求随时按照生产计划对进库原材料迅速进行优化套裁，提高材料的下料利用率。本文介绍的利用计算机代化下料方法，已在AST—386计算机上运行通过，经使用效果良好。一、优化下料的数学过程优化下料的基本数学过程是子代化→线性规划→整数规划，以此得到最优整数解。1．子优化过程设参与套我下料的零件有M种，则会有K=2M－1之多的零件组合方式，每一种组合方式又可有多个下料模式。对这些下料模式，要进行于优化，即首先通过设定一个下料分料要求，在众多的下料模式中优化出一批下料余量符合要求的下料模式。以己种零件下料为例…     

制扁弹簧下料的计算公式

根据弹簧制扁端部形状,做出几何图形,应用微积分方法对端尖体积进行计算,通过体积相等的下料原理得出弹簧下料长度的精确计算公式。并根据三种通常使用的下料方法,总结出对应的弹簧下料长度简化式,并已应用到实际生产之中。     

TYPOLOGY OF MODELS AND SIMULATION METHODS FOR MACHINING OPERATIONS

This paper presents a new modeling approach, based on Oxley's predictive model, for predicting the tool–chip contact in 2-D machining of plain carbon steels with advanced, multi-layer coated cutting tools. Oxley's original predictive model is capable of predicting machining parameters for a wide variety of plain carbon steels, however, the tool material properties and their effects are neglected in the analysis. In the present work, the effect of the tool material, more particularly, the effect of multiple coating layers and the individual coating thicknesses on the tool–chip contact length in orthogonal machining is incorporated. The results from the model predict the tool–chip contact length with respect to major cutting parameters such as feed and rake angle, work material parameters such as the carbon content in the steel, and varying thicknesses and combinations of coating layers. This model enables more precise cutting tool selection by predicting the relative tribological impact (in terms of tool–chip contact length) for a variety of multi-layer coated tools.     

AZ31B镁合金的本构方程建立及切削性能分析

基于Oxley提出的金属切削理论,通过外圆直角车削试验,利用Matlab软件进行数值迭代,逆向确认AZ31B镁合金Johnson-Cook(JC)本构方程的相关参数。将得到的JC模型参数利用自定义的方法导入到Advant-Edge分析软件中,通过比较试验和相同条件下模拟得到的切削力,最终确定本构参数,来描述加工材料的流动应力特性,优化镁合金切削加工中刀具的几何参数。     

高速硬态切削温度场和切削力动态变化的数值模拟

基于大型有限元软件ABAQUS仿真平台,建立了高速加工的有限元模型。该模型采用Johnson—Cook（JC）模型作为工件材料模型,采用JC破裂模型作为工件材料失效准则,刀-屑接触摩擦采用可自动识别滑动摩擦区和黏结摩擦区的修正库仑定律,并采用任意拉格朗日一欧拉方法实现切屑和工件的自动分离。通过有限元方法对AISI4340（40CrNiMoA）淬硬钢高速直角切削过程进行了数值模拟。通过改变刀具前角的大小,对高速硬态切削过程中刀具的温度场及切削力的动态变化进行了研究,探讨了它们各自的变化规律,研究结果有助于优化高速切削工艺,研究刀具磨损机理和建立高速切削数据库。     

AN 'EFFECTIVE CUTTING TOOL THERMAL CONDUCTIVITY' BASED MODEL FOR TOOL-CHIP CONTACT IN MACHINING WITH MULTI-LAYER COATED CUTTING TOOLS

This paper presents a new modeling approach, based on Oxley's predictive model, for predicting the tool-chip contact in 2-D machining of plain carbon steels with advanced, multi-layer coated cutting tools. Oxley's original predictive model is capable of predicting machining parameters for a wide variety of plain carbon steels, however, the tool material properties and their effects are neglected in the analysis. In the present work, the effect of the tool material, more particularly, the effect of multiple coating layers and the individual coating thicknesses on the tool-chip contact length in orthogonal machining is incorporated. The results from the model predict the tool-chip contact length with respect to major cutting parameters such as feed and rake angle, work material parameters such as the carbon content in the steel, and varying thicknesses and combinations of coating layers. This model enables more precise cutting tool selection by predicting the relative tribological impact (in terms of tool-chip contact length) for a variety of multi-layer coated tools.     

Optimizing the geometric parameters of chamfered edge for rough machining Fe–Cr–Ni stainless steel

Geometry of cutting edge has great influence on performance and reliability of modern precision cutting tools. In this study, two-dimensional finite element model of orthogonal cutting of Fe–Cr–Ni stainless steel has been built to optimize the geometric parameters of chamfered edge. A method to measure the chip curl radius has been proposed. The effect of cutting edge geometric parameters on tool stress and chip curl radius has been analyzed. Then, the chamfered edge parameters have been optimized based on numerical simulation results. It finds that, keeping the equal material removal rate, the optimal geometric parameters of chamfered edge for rough machining Fe–Cr–Ni stainless steel are that the rake angle is from 16° to 17°, and the chamfer length is from 60 to 70 μm. Small (large) rake angle combined with small (large) chamfer length is more reasonable to reduce the tool stress. When the length of land is approximately equal to undeformed chip thickness and the rake angle is larger than 15°, the chip curl radius is minimal. The groove type with large radio of width to depth should be used in the chip breaking based on the optimization results.     

OPTIMAL CONTROL OF CNC CUTTING PROCESS

０ＩＮＴＲＯＤＵＣＴＩＯＮＷｉｔｈｔｈｅｒａｐｉｄｄｅｖｅｌｏｐｍｅｎｔｏｆｅｌｅｃｔｒｏｎｉｃｔｅｃｈｎｏｌｏｇｙａｎｄｃｏｍｐｕｔｅｒｔｅｃｈｎｏｌｏｇｙ,ｔｈｅｉｎｔｅｌｉｇｅｎｃｅｏｆｍｅｔａｌｃｕｔｉｎｇｓｙｓｔｅｍｉｓｂｅｃｏｍｉｎｇｈ...     

基于Deform 3D的金属车削过程仿真

基于材料变形的弹塑性理论、热力耦合理论以及Usui磨损模型,采用有限元仿真技术,利用三维仿真软件Deform对车削过程进行仿真,得出了在不同切削用量下切削力的变化规律、切削过程中刀具中切削应力的分布情况以及刀具表面切削热的分布情况。仿真结果为生产实践中切削用量的优化选择提供理论依据。     

An intelligent fixture design method based on smart modular fixture unit

Fixture is an important assisting resource for manufacturing. Its design is under the combined influence of workpiece, machining methods, material performances, etc. and demands rich experience of the designer, which leads to significant increase in design cycle and costs. This paper researches on a new automatic fixture design methodology based on inheriting and reusing design knowledge. Traditional fixture model is changed to smart fixture model, which is not only a geometric model but also appending geometric knowledge, process planning knowledge, and performance knowledge. On the other hand, the ontology of workpiece’s machining requirement is established for driving smart fixture automatically. Based on the above ontology, the self-selection reasoning method is proposed, for machining planner, which can push unit selection scope and drive its size actively. To drive fixture unit assembly automatically, the self-assembly reasoning method is proposed according to typical feature’s assembly mode. By these ways, the geometric consistency among units as well as unit/workpiece is maintained. In the end, design efficiency and quality is expected to be improved more automatically.     

OPTIMAL CONTROL OF CNC CUTTING PROCESS

The intelligent optimizing method of cutting parameters and the cutting stable districts searching method are set up. The cutting parameters of each cutting pass could be optimized automatically, the cutting chatter is predicted through setting up the dynamic cutting force AR(2) model on-line, the spindle rotation speed is adjusted according to the predicting results so as to ensure the cutting system work in stable district.     

基于改进温度模型的300M钢本构方程参数识别

为了提高通过切削实验获取材料本构方程参数的精度,提出了将基于移动热源理论的温度分布模型沿剪切面积分计算剪切区平均温度的方法,结合不等距剪切区模型求得等效应变和应变率,建立了材料Johnson-Cook(J-C)本构方程参数的求解模型。根据切削实验获取的切削力和切屑厚度数据并采用遗传算法求得了300M钢J-C本构方程参数。与AdvantEdge FEM软件自带的300M钢本构模型相比,用所求模型参数仿真得到的主切削力、进给力和切屑厚度的精度有显著提高,验证了所建本构方程参数求解模型的有效性。