بازسازی مدل 3D سریال برای سیر تکاملی ماشینکاری بخش های چرخشی با ادغام اطلاعات برنامه ریزی عملیات معنایی و گرافیکی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|27103||2010||14 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Computer-Aided Design, Volume 42, Issue 9, September 2010, Pages 781–794
The manufacturing of a mechanical part is a dynamic evolution process from a raw workpiece to the final part, in which the generation of serial 3D models reflecting the changes on geometric shapes is especially critical to digital manufacturing. In this paper, an approach driven by the process planning course, the machining semantics and the machining geometry to reconstruct incrementally the serial 3D models for rotational part’s dynamic evolution is proposed. The two major techniques involved are: (1) extraction of machining semantics based on process planning language understanding; (2) 3D reconstruction from 2D procedure working drawings guided by machining semantics and visualization for the reconstructed series of 3D models. Compared with the conventional 3D reconstruction methods, this approach introduced the process planning course and relevant information to implement a dynamic, incremental and knowledge-based reconstruction which can greatly reduce the efforts in reconstruction and extend the collection of geometric shapes to be reconstructed.
The manufacturing of mechanical parts is a dynamic evolution process from a raw material to the final part. In this process, some changes are constantly generated on shape, function and value. With the development of digital manufacturing technology, it is highly necessary to construct the serial 3D models in relative to the part’s different manufacturing stages. The benefits of this kind of serial 3D models can be manifested on the following aspects: (1) Revealing process planning intentions. Because a part’s dynamic evolution process in geometric shape from raw material to final part is a visualized demonstration for its process planning intentions, it would be useful in the discovery, extraction, re-use and inheritance for process planning knowledge. (2) Extraction of machining features. Traditional methods extract the machining features directly from CAD model without achieving satisfactory results . Since the manufacturing process continually generates the new machining features (hole, groove and chamfer for instance), if we can construct the serial 3D models for different manufacturing stages, the machining features would be conveniently extracted in a natural way. (3) Reconstruction of part’s 3D CAD models. As the final element in above-mentioned series of part’s 3D models is corresponding with its design model, the problem to generate 3D CAD models for existing 2D-based part design results can be resolved by using this approach. (4) Generation of 3D working procedure models. The 3D working procedure models will be very useful for the complex part’s digital manufacturing such as NC programming, inspection, machining sequence verification, and 3D fixture design, etc. With 3D medial models, we can easily obtain the expected 3D working procedure models as they just correspond to some medial elements of the series. Building the serial 3D models for different machining stages in a rapid and efficient way is a challenging problem. Conventional methods for 3D model reconstruction from 2D engineering drawings have been proven to be difficult . In this paper, we propose a new approach to deal with this problem. The general idea can be explained as follows: from the point of view of dynamic process planning course by utilizing the achievement of CAPP employment, we can combine the natural language understanding with engineering drawing interpretation to get the series of part’s 3D models for dynamic machining process. The 3D reconstruction from 2D working procedure drawings will be guided by process planning describing languages which contain the valuable machining semantics. Through intelligent reasoning for the evolving process in geometric shape from raw material to final part, we can incrementally reconstruct the series of 3D models generated in part’s different machining stages. It is worthwhile to differentiate 3D working procedure models from 3D design models (i.e. 3D CAD models). 3D working procedure models are a series of geometric models which reflect the changes applied to the raw workpiece in shape and dimension during the machining process. While the 3D design model corresponds to the final state after the manufacturing is finished. For a complicated part, it will be usually necessary to add some auxiliary structures to improve its manufacturability (e.g. false bosses designed for aero-engine). The rest of this paper is organized as follows: Section 2 reviews the existing research works relevant respectively to 3D reconstruction from 2D orthographic views, natural language understanding and computer aided process planning; Section 3 describes the basic principles and proves the feasibility of our approach; Section 4 presents the key techniques used in the reconstruction process by merging semantic and graphic process planning information; Section 5 shows a case of implementation; Finally, some conclusions are given in Section 6.
نتیجه گیری انگلیسی
In this paper, a new strategy driven by process planning course and machining semantics to reconstruct the serial 3D models for dynamic evolution of rotational parts is proposed. The machining semantics involved in process planning course is defined and a machining semantics ontology is constructed with support of the static/dynamic process planning knowledge bases to represent the relevant concepts and their relationships existed in machining semantics. The machining semantics included in each WP is extracted by process planning language understanding and machining semantics ontology. An engineering drawing expressions knowledge base is established to support understanding of 2D WP (or PS) drawings. Via mapping the extracted machining semantics onto WP (or PS) drawings, the 3D WP models can be reconstructed by merging machining semantics and WP drawings understanding. Compared with the existing 3D reconstruction methods, our approach can realize a dynamic, incremental and knowledge-based reconstruction. The difficulty on 3D reconstruction can be greatly reduced by resolving the reconstruction process into multiple sub-stages (WPs) and merging machining semantics to understand 2D drawings. A case of implementation has demonstrated the feasibility and validity for this approach. It can generate, in higher automation degree, the serial 3D models to reflect workpiece’s dynamic evolution and extract 3D machining features. The serial 3D WP models obtained by our approach can be used in following digital manufacturing scenarios: a visualized exhibition, reviewing and evaluation for part’s process planning intentions and 2D-based CAPP course; and a kind of basic models for new generation of 3D-based CAPP system and for uses in NC programming, precision measuring path generation and rapid fixtures design based on 3D WP models. As our approach requires part’s machining operation sheets available, it cannot be used on design stage unlike the conventional reconstruction methods. Besides, this approach is not suitable for the sort of part which is composed mainly by freeform surfaces, as their 3D models must be given in design stage to generate the tool path by using CAM system. However, for most part types, which are composed mainly by planar surfaces and conicoid surfaces (probably with less freeform surfaces), our approach can be effectively used. In general, the 3D reconstruction principles and techniques presented in this paper can be extended to more complex part types. However, more efforts should be made to realize this expansion. The major challenges and possible solving ways could be summarized as follows: (1) 3D modeling for part’s rough material. This paper concerns only the rotational part and its rough material is generally simple (round bar stock). For some more complicated types of parts such as the box-type part, an original 3D model of rough material should be provided by casting department or generated by using a CAD system. (2) Determination of positional dimensions or benchmarks for current machining features. For more complex part types, as their neighboring WPs drawings could be greatly diverse unlike in the rotational part case, the existing machining features or benchmarks information are often incomplete or indirectly represented in the current WP drawings, it will lead the positioning of some complex machining features more difficult. To overcome this difficulty, a feasible way is to utilize the precursory 3D WP model. As the precursory 3D WP model is an all-information model for latest reconstruction result and it contains complete shape and positional information for existing machining features. Moreover, the precursory 3D WP model can conveniently generate 2D project view along arbitrary specified direction. So, we can generate on-demand some supplementary project views by using the precursory 3D WP model to establish an association on shape and position between the current WP drawings and the precursory 3D WP model. In this way, the positional dimensions or benchmarks for current machining features could be effectively determined. (3) More perfect machining semantics’ definition, extraction and mapping. For more complicated types of parts, more abundant machining semantics and relationships existed among them have to be considered and analyzed. The knowledge bases for process planning language understanding and engineering drawing expressions should be further extended. Meanwhile, some appropriate human–computer interactions could be introduced. As an original work, this paper primarily concerns the serial 3D WP models reconstruction for rotational part. In our future research, we will extend our approach to more complicated part type such as the prismatic part with milling and other features to further verify and improve the techniques and methods proposed in this paper.