Tool chatter prediction based on empirical mode decomposition and response surface methodology

•Microphone is an appropriate sensor for acquiring chatter signals.•The recorded chatter signals are non-stationary, nonlinear in nature.•Empirical mode decomposition has the ability to filter out the contaminations from the recorded chatter signals.•From the results it has been found that, depth of...

Full description

Saved in:
Bibliographic Details
Published in:Measurement : journal of the International Measurement Confederation 2021-03, Vol.173, p.108585, Article 108585
Main Authors: Shrivastava, Y., Singh, B.
Format: Article
Language:English
Subjects:
Background noise
Chatter
Correlation analysis
Decomposition
Empirical analysis
Empirical mode decomposition
Intrinsic mode function
Machining
Mathematical models
Noise
Parameters
Prediction models
Response surface methodology
Signal processing
Studies
Vibration
Vibration analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Microphone is an appropriate sensor for acquiring chatter signals.•The recorded chatter signals are non-stationary, nonlinear in nature.•Empirical mode decomposition has the ability to filter out the contaminations from the recorded chatter signals.•From the results it has been found that, depth of cut and cutting speed have more effect on the chatter severity. Despite the immense work done within the domain of vibration and noise mitigation, still many features related to chatter remains unexplored. Researchers had suggested techniques to explore the mechanism of tool chatter based on acquired chatter spectrum. However, the effect of background noise have been overlooked. In present study, recorded chatter signal have been pre-processed using empirical mode decomposition (EMD). Thereafter, obtained intrinsic mode functions (IMFs) have been subsequently analyzed to identify the most dominating IMF pertaining to tool chatter. Further, a new parameter chatter index (CI) has been evaluated for each corresponding IMFs as a response. Moreover, prediction model has been generated using response surface methodology, in order to establish the dependency of tool chatter on various machining parameters. Finally, more experiments have been conducted to validate the aforesaid chatter severity model. Well correlation between the experimental results and predicted ones authenticates the proposed technique.
ISSN:0263-2241
1873-412X
DOI:10.1016/j.measurement.2020.108585