برنامه ریزی عملیات مبتنی بر ویژگی هوشمند برای بخش های آسیاب چرخشی پنج محور

This paper proposes a new process planning system for five-axis mill-turn parts. This system focuses on parts including both prismatic and rotational features newly defined as Prisronal parts. The three main parts of the new system are as follows: (a) a novel machining features classification based on machining processes and number of simultaneously controlled axes of five-axis lathe; a new group code system is proposed. (b) Machining feature definition with representation; this model not only includes geometric and manufacturing data in the open layer but also includes machining processes and machining knowledge viewed as production rules of expert systems. (c) Process plan generation model; a process plan is generated based on machining features and rules. A user interactively identifies geometric entities to machining features provided as templates by the system. Machining processes and particular machining rules are designed and built-in each machining feature in the hidden layer. Auxiliary rules are provided for setup selection, precedence relation, and tool selection. Benefits of the system are fast and accurate process plan generation.

Process planning is a decision-making process. It is done normally by a human and depends on individual experience. A feasible and efficient solution requires a proficient process planer. Process plans made by a human expert are inconsistent, inaccurate, time-consuming and costly Elias [1]. Therefore, an expert system could be a good choice for CAPP. This paper presents a new intelligent expert process planning system for five-axis mill-turn parts. Three important problems need a solution for developing this system. The first is representation of workpiece geometry and technological data. The second is identifying machining processes and knowledge-based rules. The third is generating a process plan. To overcome the first and second problems, Feature concepts are applied Shah et al. [2] and [3]. To solve the last problem, machining processes, knowledge-based rules and heuristics are designed and attached to each feature. The system is based on variant and generative approach. The variant concept is applied in the sense of the machining features classification and a group code system whereas the generative concept is applied in term of rules for decision-making processes. This study is distinctive from previous CAPP research by focusing on prisronal parts. “Prisronal part” is the author's definition. It is defined as “a class of parts having primitive shapes with one common centerline like cylindrical, cone or round shapes including other symmetric and/or asymmetric rotational machining features for example holes, faces, slots, grooves or pockets in different orientations”. Prisronal parts are usually found in automotive, ship and aerospace industries such as shafts, propellers and impellers. The paper is structured as follows. Section 2 gives literature review. Section 3 shows the four components of feature-based process planning system: design of system architecture, machining features classification, a new group code system, and machining feature definition and representation. Section 4 shows process plan generation model. Section 5 shows a case study. Finally, conclusions are given.

A new intelligent process planner for mainly four and five axis mill-turn parts was proposed. These mill turn parts are newly defined as Prisronal (prismatic and rotational). A new machining features classification based on machining processes and number of simultaneously controlled axes of five-axis lathe was proposed, Although, the machining features can be machined separately by a simple lathe and five-axis milling machine, prisronal parts can be produced faster and more accurate on a five-axis lathe. A new group code system combining the variant and generative approaches is designed to define prisronal parts. This code includes a feasible sequence of machining and a hierarchy to model interacting features. The machining feature definition model and representation model includes machining processes and knowledge for machining each feature. Machining feature representation is designed to guide user to interactively define features based on templates. Process plans are generated automatically by knowledge based approach in three modules: (i) the hidden layer of machining features, attached machining processes, rules and heuristic algorithms, (ii) the advising module including an algorithm for advising alternative milling features and (iii) the support module containing 50 auxiliary rules for four general procedures: setup selection, precedence relation between operations, selection of cutting tools and tool path generation. The output is kept in the STEP files as comments to allow interface with tool path generation in commercial CAM systems. The benefits of the system are: (1) fast and accurate process plan, (2) group code and corresponding processes generated by the system can be archived and used as a variant process planning system