CN-MgMP: a multi-granularity module partition approach for complex mechanical products based on complex network

Module partitioning is beneficial for engineers to gain a better understanding of the structure and function of complex mechanical products (CMPs). It plays an important role in the entire life cycle of the CMPs. However, owing to the large number of mechanical parts and the complex relationships, e...

Full description

Saved in:
Bibliographic Details
Published in:Applied intelligence (Dordrecht, Netherlands) Netherlands), 2023-07, Vol.53 (14), p.17679-17692
Main Authors: Zhang, Zhenjie, Lu, Botao, Xu, Xiaobin, Shen, Xufeng, Feng, Jing, Brunauer, Georg
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Clustering
Computer Science
Heuristic
Integer programming
Machines
Manufacturing
Mathematical models
Mechanical Engineering
Modularity
Modules
Optimization
Processes
Product design
Semantics
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Module partitioning is beneficial for engineers to gain a better understanding of the structure and function of complex mechanical products (CMPs). It plays an important role in the entire life cycle of the CMPs. However, owing to the large number of mechanical parts and the complex relationships, existing module partition approaches cannot obtain multi-level information. Consequently, the in-depth analysis of CMPs is hindered. Therefore, a novel multi-granularity module partition approach for CMPs based on a complex network (CN-MgMP) is proposed in this paper. Using this approach, a weighted complex network for mechanical parts (MP_WCN) was constructed. The multi-granularity module partition algorithm is presented based on MP_WCN. It consists of three phases. First, based on the concept of modularity increment, the MP_WCN is decomposed into several sub-networks, and fine-grained modules are thus obtained. Next, the topic semantic vectors of these fine-grained modules are extracted using the termfrequency–inversedocumentfrequency (TF-IDF) text analysis method. Finally, taking the global module distance (Glo-MD) as the optimization goal, the fine-grained modules are merged hierarchically to obtain optimal coarse-grained modules. The experimental results for the vertical elevator demonstrate the effectiveness and superiority of the proposed approach.
ISSN:0924-669X
1573-7497
DOI:10.1007/s10489-022-04430-2