پیاده سازی برنامه ریزی عملیات غیر خطی مبتنی بر ISO 14649 برای ماشینکاری پیچیده

Increasing attention is being paid to complete machining, i.e., machining of the whole part in a single machine tool, in the metal working industry. For this purpose, complex machine tools equipped with machining components, such as multiple spindles and turrets have been developed by leading machine tool builders. The efficiency of complex machine tools is largely dependent on how the machining components are utilized. The main thrust of this paper is twofold: (1) Proposition of a nonlinear process planning based on the STEP-NC (STEP-compliant data interface for numerical controls) paradigm whose data model is formalized as ISO 14649, and (2) Development of an optimal solution algorithm for process planning for complex machining. The developed algorithm is based on the branch-and-bound approach and heuristics derived from engineering insights. The developed process planning method and optimization algorithm were implemented and tested via the TurnSTEP system developed by our research team. Through the experiments, we are convinced that the new process planning and algorithm can be used as a fundamental means for implementing the third type of STEP-NC [Suh S. TurnSTEP: Tools to create CNC turning programs. In: White paper presented on STEP Implementers’ Forum ISO TC184/SC4 Meeting. 2004], i.e., an Intelligent and Autonomous STEP-NC system for the CAD-CAM-CNC chain supporting e-Manufacturing.

With the advancement of machine tool building and computer numerical control (CNC) technology, various configurations of CNC machine tools have been undergoing continuous development. In the typical machine tool scenario, machine tools with as many as 20 mechanical axes with 10 simultaneous controls are commonly seen. Using highly sophisticated machine tools, manufacturing productivity can be greatly enhanced because the entire machining operation required to fabricate the part can be made in a single setup in the same machine tool. For instance, the whole part can be machined without off-loading by using dual spindles, while the machining time can be greatly reduced and by using multiple turrets. Moreover, both turning and milling operations can be done in the same machine tool. Specifically, complex machining is defined as a set of machining operations that can be typically made by multi-channel complex machine tool (MCCM), equipped with more than one spindle and turrets for both turning and milling operations. MCCM has been developed by the machine tool makers in response to industrial demands for complete machining of complex parts with single machine tools. For example, in an MCCM equipped with 2 spindles and 2 turrets, the upper turret has four degrees-of-freedom of motion: three (X,Y,Z)(X,Y,Z) for translation and one (B) for rotation. Together with the CC-axis motion from the spindle, the machine can perform 5-axis milling operations. The lower turret can be used for turning of the parts attached to the two spindles. Further, this type of machine tool is normally equipped with part loading and unloading mechanisms. Compared with conventional machine tools such as turning machines, for instance, a number of advantages can be obtained. (1) Complete machining: In one setup of the workpiece, all operations required for the workpiece can be made by MCCM. (2) Automated setup operations: The whole operations from loading of the workpiece to unloading after complete machining can be automated. (3) One-feature simultaneous machining (OFSM): The workpiece can be machined by more than one cutting tool to reduce the cycle time, i.e., the same removal volume (formalized as turning_machining_feature in ISO 14649-12 [1]), is machined by more than one cutting tool, as shown in Fig. 1(a). (4) Two-feature simultaneous machining (TFSM): The cutting tools remove different turning_machining_features [1], as shown in Fig. 1(b), to reduce the cycle time. (5) Parallel machining (PM): Distinguished from the two simultaneous machining cases, two workpieces attached to the two spindles may be machined by more than two turrets, as shown in Fig. 1(c). Full-size image (18 K) (a) One-feature simultaneous machining. Figure options Full-size image (18 K) (b)Two-feature simultaneous machining. Figure options Full-size image (19 K) (c) Parallel machining. Figure options Fig. 1 Three feasible machining modes in MCCM. MCCM’s great potential is not fully realized in practice, mainly due to inefficient process planning methods for complex machining with MCCM. Process planning herein means micro-process planning (dealt with in ISO 14649) involving a single type of machine tool, not macro-process planning (dealt with in ISO 10303 AP240) involving more than one machining process with more than one machine tool, e.g., Kruth [2]. In most industrial practices, process planning and part programming are done manually by skilled operators, or by computer-aided systems installed either in off-line CAM systems or in on-line CNC controllers, the so-called CAPS (Conversational Automatic Programming System). Moreover, with CAPS, the current degree of automation and optimization achievable is not sufficient. Commercial CAM systems do not offer nonlinear process planning or algorithms for synchronous machining, but do provide a suitable environment for users to select a synchronous machining mode for each turret and specify cutting tools, removal volumes, and process attributes. In the literature on CAPP, Elmaraghy [3] summarized relevant research including future prospects. Specific problems of process planning for MCCM have not been dealt with much in the previously reported open literature. The previous literature related with micro-process planning of turning and complex machining operations can be summarized as follows. For turning operations, Zang [4] proposed a knowledge-based, feature recognition method called EXCAP, and Joseph [5] and Kalta [6] and [7] expanded the functionality of EXCAP system by adding CAM and CAD functions. Barakat [8] developed a variant CAPP system, where a process plan is generated by modifying a process plan for a similar part. The above CAPP systems are subject to conventional turning machines and have a limitation in not supporting complex machining. A group of researchers including Levin [9] and [10], Yip-Hoi [11] and [12] suggested algorithms for finding the delta volumes to be removed by turning for turn-mill parts and for generating the schedule for parallel machining. In their work, complex machining is limited to parallel machining. Note that complex machining is not necessarily the same as parallel machining where only one turret and only one spindle are utilized for complete machining, and hence their algorithms cannot be applied to the complex machining problems including simultaneous machining addressed in this paper. For complex machining, only a few works have been carried out. Chiu [13] proposed a genetic algorithm for scheduling a machining sequence. Veeramani [14] suggested a hybrid CAPP system, where the process plan is generated by means of computer algorithm and user interaction. As far as the paradigm of process planning is concerned, all previous research assumed that their (optimal) solutions found off-line are given to a CNC controller via machine tools and control-specific code, according to G-code (ISO 6983). Due to the nature of G-code, the solution has to be represented as a linear process plan, i.e., a sequential process made of NC blocks. However, this is not desirable in two respects. (1) In converting CAPP solutions to G-code format, some delicate information, such as one/ two-feature simultaneous machining in MCCM cannot be represented. (2) Even if they are precisely represented by a part program, the part program is fixed information to be executed by the CNC controller, allowing no room for flexibility. In the case where the assumed environment in the process planning stage cannot be maintained on the shop floor due to some event (e.g., unavailability of the cutting tools, tool wear/breakage, etc.), the optimality cannot be guaranteed or the machining operation cannot be completed. Further, most of the previous research did not take inter-operability into consideration. In this paper, we present a nonlinear process planning method based on STEP-NC paradigm whose data model is formalized as ISO 14649 emerging as the new interface between CAM and CNC. In particular, we address the structure of process planning system capable of nonlinear process planning, and develop an optimization algorithm for determining process sequences and time schedule for complex machining so that the total cycle time is minimized. The paper is organized as follows. In Section 2, the overall structure of the STEP-NC-based process planning system capable of nonlinear process planning is presented. To support such a system, a brief background on STEP-NC and a new data model for the new process planning method are given. Among the various algorithms which need to be developed to implement such a process planning system, solution algorithm for determining an optimal process sequence (called EPSG: Executable Process Sequence Graph) defined in Section 3 is derived in Section 4. The nonlinear process planning and algorithm are validated via experiments with the TurnSTEP system developed by our research team in Section 5. This paper is concluded with a few remarks in Section 6.

In this paper, we addressed two issues: (1) Nonlinear process planning based on the STEP-NC paradigm whose data model is formalized as ISO 14649, and (2) Optimal process planning for complex machining. Taking complex machining as an example, where the effect of nonlinear process planning is accentuated, we proposed a nonlinear process planning composed of NPSG, HPSG, and EPSG. To fully appreciate the new process planning, optimization of EPSG is of practical importance. For such a purpose, we developed an optimization algorithm minimizing the total cycle time based on the branch-and-bound approach, and heuristics derived from engineering insights. The implementation of the developed algorithm on the TurnSTEP system ensures that the developed algorithm can be used for industrial practice in micro-process planning for complex machining with MCCM, even for a non-STEP-NC environment. This is because, although the input used in this algorithm is based on the STEP-NC data model (ISO 14649), it does not necessarily bind the application environment. In summary, the new process planning and algorithm for EPSGs developed in this paper can be used as the fundamental tools for implementing the third type of STEP-NC [20], i.e., an Intelligent and Autonomous STEP-NC system for the CAD-CAM-CNC chain supporting e-Manufacturing.