Decision rule mining for machining method chains based on rough set theory
- PDF / 559,606 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 13 Downloads / 185 Views
Decision rule mining for machining method chains based on rough set theory Rui Wang1 · Xiangyu Guo2 · Shisheng Zhong2 · Gaolei Peng1 · Lin Wang1 Received: 1 April 2020 / Accepted: 5 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Decision rules for machining method chains mined from historical machining documents can help technologists quickly design new machining method chains. However, the main factor that limits the practical application of existing rough set models is that the boundary regions are too large. Therefore, a decomposition-reorganization method (DRM) is proposed to mine rules for machining method chains. First, binary coding is used to decompose the existing machining method chains, and the decision rules for a single machining method are mined based on rough set reduction. Then, machining method chains are obtained by reorganizing the machining methods in accordance with the decision rules. DRM can eliminate the boundary regions without human intervention and recommend machining method chains for all features whose parameters have appeared in historical machining documents. Finally, three types of shell parts are used to verify the effectiveness of DRM. Keywords Machining method · Process planning · Rule mining · Rough set theory · Derivation rules
Introduction For convenience of elaboration, several specially defined terms are used in this paper. First, feature parameters refer to both dimensional parameters (such as length and diameter) and machining requirements (such as precision requirements and rugosity requirements). Second, most features cannot be machined in a single step (Xu et al. 2020; Chhim et al. 2019); for example, a hole with a precision of the 6th grade requires three machining steps: “drilling → expanding → reaming”. Therefore, for clarity, a machining method refers to a single machining step (such as drilling), whereas a machining method chain refers to the entire machining process composed of multiple machining methods for one feature (such as “drilling → expanding → reaming”). The selection of machining method chains is very important. An incorrect machining method chain may cause the machined parts to need to be scrapped or reworked (Milica
B
Lin Wang [email protected]
1
School of Ocean Engineering, Harbin Institute of Technology, No. 2 Wenhuaxi Road, Weihai 264200, China
2
School of Mechatronics Engineering, Harbin Institute of Technology, No. 92 West Dazhi Street, Harbin 150000, China
et al. 2016; Ye et al. 2020). In practical engineering, technicians will spontaneously recall or consult similar machining cases that have been addressed before to select a suitable machining method chain to use, with or without modification (Li et al. 2020; Oztemel and Gursev 2020). Therefore, the main research content of this paper is to simulate the reasoning process of the human brain. Through an analysis of historical cases, the decision rules relating the feature parameters to the machining method chains can be mined, which can assist
Data Loading...