A complex network based NC process skeleton extraction approach

•The characteristics and conditions for a good process skeleton are analyzed and evaluated to guide the extraction of process skeleton.•The structured model for representing typical macro process based on CAM model data is elaborated for finer discovery of macro process knowledge.•A NC machining pro...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers in industry Jg. 113; S. 103142
Hauptverfasser: Huang, Bo, Zhang, Shusheng, Huang, Rui, Li, Xiuling, Zhang, Yajun, Liang, Jiachen
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.12.2019
Schlagworte:
CAM model
Complex network
Macro process
NC process reuse
Process skeleton
NC process reuse
CAM model
Process skeleton
Macro process
Complex network
ISSN:0166-3615, 1872-6194
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•The characteristics and conditions for a good process skeleton are analyzed and evaluated to guide the extraction of process skeleton.•The structured model for representing typical macro process based on CAM model data is elaborated for finer discovery of macro process knowledge.•A NC machining process data network by using the machining process instances is constructed to judge the critical of process element.•The final process skeletons are determined by choosing those candidate process skeletons with moderate granularity and conforming to the specific process. Process skeleton can effectively reflect the macro process constraints and key features of parts, which has rich reuse value. However, neither the automatic extraction of common reusable structures at geometric level nor the discovery of typical process routes at semantic level can effectively support the reuse of macro processes. To counter this problem, a novel complex network based NC process skeleton extraction approach is proposed. First, the characteristics for a good process skeleton are analyzed and the structured model for representing typical macro process based on CAM model data is elaborated. Then, given a set of structured NC machining process instances, the corresponding extraction approach is proposed, which includes three important phases: (1) constructing a NC machining process data network by using the machining process instances; (2) analyzing the topological metrics of the process data network to judge the key process element nodes and generating candidate process skeletons; (3) determining the final process skeletons by choosing those candidate process skeletons with moderate granularity and conforming to the specific process. Finally, the experimental results are presented to demonstrate the effectiveness of the approach. The presented approach can help smart machines reuse the machining know-how about cutting similar parts so that parameter settings can be updated automatically.
ISSN:0166-3615
1872-6194
DOI:10.1016/j.compind.2019.103142