Efficient stability prediction of milling process with arbitrary tool-holder combinations based on transfer learning

Congying Deng (), Jielin Tang (), Jianguo Miao (), Yang Zhao (), Xiang Chen () and Sheng Lu ()
Additional contact information
Congying Deng: Chongqing University of Posts and Telecommunications
Jielin Tang: Chongqing University of Posts and Telecommunications
Jianguo Miao: Sichuan University
Yang Zhao: Chongqing University of Posts and Telecommunications
Xiang Chen: Chongqing University of Posts and Telecommunications
Sheng Lu: Chongqing University of Posts and Telecommunications

Journal of Intelligent Manufacturing, 2023, vol. 34, issue 5, No 11, 2263-2279

Abstract: Abstract Chatter occurring in the milling process can seriously deteriorate the machining efficiency and surface quality. The stability diagram predicted using the tool tip frequency response functions (FRFs) is an effective approach to avoid the chatter vibration. The tool tip FRFs highly depend on the characteristics of the tool-holder-spindle-machine tool frame assembly. Thus, when the tool-holder assembly or only the tool overhang length changed, the FRFs will be reobtained to plot the stability diagrams. Considering this time-consuming situation, this paper introduces the transfer learning to efficiently predict the milling stability of arbitrary tool-holder combinations. First, a source tool-holder assembly is selected to measure sufficient overhang length-dependent tool tip FRFs and then predict the limiting axial cutting depth aplim values under different process parameters for forming the source data. For a new tool-holder assembly, impact tests are only performed under a few key tool overhang lengths to measure the tool tip FRFs and then predict the aplim values to form the target data. Combining the target data and the source data, the transfer learning containing the domain adaptation and adaptative weighting is introduced to train an overhang length-dependent milling stability prediction model of a target tool-holder assembly. A case study has been performed on a vertical machine tool with four different tool-holder assemblies to validate the feasibility of the proposed transfer learning-based milling stability prediction method.

Keywords: Milling stability; Tool-holder assembly; Transfer learning; Tool overhang length (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s10845-022-01912-5 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:34:y:2023:i:5:d:10.1007_s10845-022-01912-5

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-022-01912-5

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().