基于STEP-NC的工艺规划优化方法研究

在重点分析零件特征典型加工方案及加工方案组合优化原则的基础上,基于STEP-NC应用遗传算法建立了零件加工方案组合优化的数学模型。并将该方法应用到具体零件的工艺排序决策过程中,通过编码、杂交、复制、变异等得到满足零件要求的最优或接近最优的工艺路线。验证了遗传算法在STEP-NC工艺路线排序中的有效性。     

典型零件磨削加工工艺智能决策系统

磨削作为大多数典型零件终加工方法,开展典型零件磨削加工工艺智能决策是满足其高精、高效的重要手段,因此提出基于6R模型的典型零件磨削加工智能工艺决策系统框架。首先,结合实例推理与置信度计算各实例的综合置信因子开展实例优选。然后,建立异质集成学习与粒子群算法优化的预测模型获得最优工艺方案输出。最后,基于Qt 4.8.7和SQLite 3开发典型零件磨削加工工艺决策系统。以磨床砂轮主轴磨削为例,根据企业应用报告,使用工艺智能决策系统的方案加工后,零件表面质量提高73.25%,工艺方案决策时间从近20 h缩短到2～4 h,零件加工处理时间从10 min缩短到5 min,单个产品的加工效率提高25%以上,最高达47%,工艺决策准确率达到98.2%,实现了典型零件的高效、高精度磨削加工。     

基于BP神经网络的槽类特征铣削加工方案决策和刀具优选

为实现CAPP系统中零件特征加工方法的智能决策,完成加工结果相互关联的特征加工方案选择,利用BP神经网络算法对槽类特征加工方法和刀具选择进行智能决策,建立了从输入到输出的网络决策模型,并对网络结构、参数确定问题进行阐述。通过正交表选取试验样本,运用带有自适应lr的梯度下降算法来训练网络,最后通过实例验证了该网络的有效性。结果表明,利用BP神经网络进行零件加工方法的选择和刀具优选是可行且有效的。     

基于灰色关联分析的加工方案决策研究

机械零件尽管多种多样,但均是由一些诸如外圆、内孔、锥面、平面、螺纹、齿形等常见形状特征所组成。因此,合理选择常见形状特征的加工方案,是正确制定零件加工工艺的基础。提出了基于特征的二级加工方案决策体系,建立了基于零件特征的加工方案评价指标体系,并采用灰色关联分析对特征加工方案进行优化分析和计算。最后以单特征圆柱孔的工艺方案决策为例,证明了该方法的实用性。     

箱体零件工艺路线分层决策方法

箱体零件特征面多，加工方法多，工艺路线决策的规模大，决策算法复杂。文章将工艺路线决策过程分解成特征工艺决策、加工路线排序、工序决策、工步决策四个层次。决策过程中，将决策算法与零件的具体形状特征、加工方法相分离，简化了算法，提高了决策算法的实用性和灵活性。     

基于模糊推理的零件特征加工方案决策研究

为实现计算机辅助工艺设计(Computer Aided Process Planning,CAPP)系统中零件特征加工方案的决策,分析了影响加工方案选择的因素,提出了一种基于模糊推理的零件特征加工方案决策方法。文中建立了零件特征属性与加工方案的模糊映射关系,设计了适当的隶属度函数,建立了加工方案的模糊推理模型。该决策方法将基于规则的推理转化为模糊推理,降低了推理过程的匹配冲突。通过实例验证表明,利用模糊推理进行零件特征加工方案的决策是有效的和可靠的。     

基于模糊BP网络的加工方法链决策

为实现计算机辅助工艺计划(CAPP)系统中零件特征加工方法的决策,分析了影响零件加工方法的因素,提出利用模糊BP神经网络算法对零件特征加工方法链进行优化决策.根据BP网络能够学习任何模式的映射关系的特点,建立了从输入到输出的网络决策模型,并对网络结构、参数确定和样本选取等问题进行详细阐述,运用合理的学习算法来训练网络,最后通过实例验证了该网络的有效性.结果表明,利用模糊BP神经网络进行零件加工方法的选择是有效的和可靠的.     

箱体零件数控加工路线研究

针对箱体类零件的特点,探讨了箱体零件加工工艺路线的确定原则和程序;阐述了箱体类零件的特征型面描述方法;建立了不同特征的加工方法匹配规则库,实现了零件特征要素加工方法的决策;同时研究了零件一次装夹的优化问题,建立了数学模型.实现箱体零件特征要素加工方法的决策和工步排序,最终得到了箱体零件在加工中心上的最优工艺路线.对传统工艺过程进行优化具有非常重要的工程实用价值.     

基于仿真的数控铣削加工参数优化研究

针对数控铣削加工参数优化问题,通过加工仿真,计算每一走刀步的切削深度和切削宽度并划分区间。在每一个由组合划分区间内所有刀步构成的一个加工特征组合段上建立了多目标优化模型,模型采用遗传算法对每一加工特征组合段的加工参数进行优化,自动修改NC程序中的加工参数并反映优化结果。应用实例证明,提出的多目标优化模型和优化求解算法正确、有效。     

孔群加工路径规划问题的进化求解

孔群加工路径规划对于提高多孔类零件的加工效率和质量具有重要意义。建立了两个孔群加工路径规划问题的数学模型,分别归纳为单目标和多目标组合优化问题,并引入进化蚁群系统算法和人工免疫算法求解单目标组合优化问题。这两种算法均能有效防止解空间的“组合爆炸”问题,计算复杂度的阶次低于Hopfield神经网络算法,且性能优于Hopfield算法。采用多目标解的快速排序技术分别对进化蚁群系统算法和人工免疫算法加以改进,开发出多目标进化蚁群系统算法和多目标人工免疫算法。分析表明,改进算法不增加原算法的计算复杂度,能直接用于求解多目标组合优化问题而无需事先给出目标权值向量,并能一次运行求得问题的多个Pareto最优解。     

Algorithms for computer-aided generative process planning

Process planning is the function within a manufacturing facility that establishes the machining processes and parameters to be used so as to convert a piece-part from its initial form to the final form predetermined on an engineering drawing. Computer-aided process planning (CAPP) has become a major focus of manufacturing automation as it forms the interface between computer-aided design (CAD) and computer-aided manufacturing (CAM). Issues in CAPP include part representation, process selection, alternative process-plan generation, intermediate surface and tolerance determination, and operation sequencing. This paper focuses on quantitative models for determining cutting dimensions and tolerances for intermediate surfaces, and on a heuristic for sequencing cutting operations.     

A Dynamic Programming Based Process Planning Selection Strategy Considering Utilisation of Machines

Computer-aided process planning (CAPP) has been recognised as a key element in computer-integrated manufacturing (CIM). Despite the fact that tremendous efforts have been made to develop CAPP systems, the benefits of CAPP in real-life manufacturing environments are not obvious. One of the main difficulties in applying CAPP in real manufacturing settings is the missing link between generated process plans and shop-floor activities. The assumptions that the shop floor remains static and there are unlimited resources available have caused a typical CAPP system to make a repetitive selection of the same most desirable machines for manufacturing different parts. When process plans for various parts are finally sent to the shop floor, process bottlenecks normally arise, making the generated plans ineffective. Some efforts have been made in the integration of process planning and shop floor activities; however, the problems associated with the lengthy time and impractical attempts at the generation of process plans, remain unsolved. Since in most cases, it is possible that several alternatives exist for machining a part by using different equipment, a stage-type network structure can be constructed. This paper introduces a dynamic programming based process plan selection strategy, enabling the efficient solution of this stage-type network prob-lem. In this proposed strategy, the use of workstations is also taken into consideration so that the resulting plan is valid for the conditions that exist on the shop floor at the time an order is released. A case study has been provided to substantiate the effectiveness of this proposed strategy.     

A hybrid approach for set-up planning

In this paper, various constraints on set-up planning are identified and discussed. Rules and heuristics are created based on these constraints to determine the bilateral precedence relationships between the machining process of features and tool approach directions. By looking at the problem caused by the conflict between the feature precedence relationships and set-up precedence relationships, an optimisation approach is developed to find the optimal plan(s) from all the feasible set-up candidates. The criterion used is the minimum number of set-ups. As a result, alternative set-up plans are generated with a certain number of set-ups required. This research is limited to machining prismatic components only.     

On-line integration of a process planning module with production scheduling

An approach to the on-line integration of process planning and production scheduling is reported. Based on a geometric modeller, a geometric analyser and a knowledge base, the process planner generates alternative process plans and provides automatic tool selection and calculation of the appropriate machining parameters. Time and cost estimations are input to the decision-making module in the production scheduling system that produces optimal scheduling decisions as well as a complete record of the actual state of the factory resources. An information flow, designed as a relational data model, maintains the interaction between the process planning and the production scheduling systems and provides the dynamic feedback to the process planner. Specific geometry features of the blank, the finished part and the cutting tools, and geometry features-based rules are stored in the database of a developed CAPP system by using a separate CAD interface. The integrated production planning and production scheduling system and the CAPP system were validated with rotational parts machining.     

A new accurate curvature matching and optimal tool based five-axis machining algorithm

Free-form surfaces are widely used in CAD systems to describe the part surface. Today, the most advanced machining of free from surfaces is done in five-axis machining using a flat end mill cutter. However, five-axis machining requires complex algorithms for gouging avoidance, collision detection and powerful computer-aided manufacturing (CAM) systems to support various operations. An accurate and efficient method is proposed for five-axis CNC machining of free-form surfaces. The proposed algorithm selects the best tool and plans the toolpath autonomously using curvature matching and integrated inverse kinematics of the machine tool. The new algorithm uses the real cutter contact toolpath generated by the inverse kinematics and not the linearized piecewise real cutter location toolpath.     

基于QFD的数控加工工艺质量优化规划方法

提出了一种基于质量功能配置的确定工艺质量水平的建模方法,并建立了优化数学模型,其目标函数是总体工艺质量改进程度最大化,约束条件包括三个方面：①工艺质量水平变化区间。②工艺质量要素之间的相关性。③预算的工艺质量成本。应用实例表明,针对具体的工艺质量问题,运用该模型能够帮助工艺设计人员有效地规划出最优的工艺质量水平。     

Towards the integration of process and production planning: an optimisation model for cutting parameters

This paper aims to contribute to process and production planning integration through the development of a new cutting parameters optimisation model. The developed model considers simultaneously the technology-related constraints and a shop floor constraint determined by the available time at each workstation. The latter, being a constraint related to the part machining time, is associated with the set of all elementary machining operations and implies the development of a new multi-operation optimisation model. In this approach, part machining time is a new variable for shop floor scheduling. Since the limiting factor of workstation available time at every scheduling date depends on the shop floor status, optimum part machining time can range from the time for minimum cost to the time for maximum production rate. The introduction of the available time in the optimisation process allows for the generation of improved schedules according to several performance measures. The proposed optimisation model is non-linear, uni-criterion and multi-variable. The search of the optimal solution is carried out using sequential quadratic programming.     

Machinability and set-up orientation for five-axis numerically controlled machining of free surfaces

In this paper, we investigate the problems of determining machinability and part set-up orientation for a given surface model. We first develop kinematic models of the five-axis machines, based on the axis configuration, then develop algorithms for determining the feasibility of one-set-up machining (machinability) and the part orientation for types C-A and A-B tool configurations. Machinability is determined by a computationally efficient procedure for finding the intersection of the feasible tool motion on the sphere and a visibility map which we call the binary spherical map (BSM). The part set-up is so chosen as to give the smallest rotational range among all feasible configurations. The algorithms developed have been tested via numerical simulations. The results show that they can be readily implemented in a CAD/CAM system, as an automated process planner selecting the efficient machine type and part set-up for NC machining.     

GA-based synthesis approach for machining scheme selection and operation sequencing optimization for prismatic parts

To obtain global and near-global optimal process plans based on the combinations of different machining schemes selected from each feature, a genetic algorithm-based synthesis approach for machining scheme selection and operation sequencing optimization is proposed. The memberships derived from the fuzzy logic neural network (FL-NN), which contains the membership function of each machining operation to batch size, are presented to determine the priorities of alternative machining operations for each feature. After all alternative machining schemes for each feature are generated, their memberships are obtained by calculation. The proposed approach contains the outer iteration and nested genetic algorithm (GA). In an outer iteration, one machining scheme for each feature is selected by using the roulette wheel approach or highest membership approach in terms of its membership first, and then the corresponding operation precedence constraints are generated automatically. These constraints, which can be modified freely in different outer iterations, are then used in a constraints adjustment algorithm to ensure the feasibility of process plan candidates generated in GA. After that, GA obtains an optimal process plan candidate. At last, the global and near-global optimal process plans are obtained by comparing the optimal process plan candidates in the whole outer iteration. The proposed approach is experimentally validated through a case study.     

[工艺规划与装配线平衡]两阶段混合算法求解集成工艺规划与调度问题

提出了求解集成式工艺规划与车间调度问题的两阶段混合算法。在工艺规划阶段,使用遗传算法为每个工件生成可选的近优工艺路线集,动态地为车间调度阶段输入已确定的工艺路线;在车间调度阶段,使用蜜蜂交配优化算法快速寻优,设计了蜂王婚飞的流程以保证算法的全局搜索能力,构建了基于不同邻域结构的工蜂培育幼蜂局部搜索策略。使用基准测试集对提出的方法进行验证,并与现有算法进行对比,计算结果证明了提出方法的有效性。