Tool wear monitoring based on kernel principal component analysis and v-support vector regression

Abstract

Machined surface quality and dimensional accuracy are significantly affected by tool wear in machining process, and severe tool wear may even lead to failing of the workpieces being processed. Tool wear monitoring is highly desirable to realize automated or unmanned machining process, which can get rid of the dependence on skilled workers. This paper mainly studies on the methods and techniques of on-line tool wear monitoring through static and dynamic cutting force signals. Sensitive signals related to tool wear are preliminarily selected by using correlation coefficient method. Kernel principal component analysis (KPCA) technique is adopted to fuse these sensitive features for improving training speed and prediction accuracy. Then, the tool wear predictive model based on v-support vector regression (v-SVR) is constructed through learning correlation between the fused features and actual tool wear. The obtained result shows that the prediction accuracy of the proposed tool wear model is proved effective beyond expectation. Besides, the proposed model still has better generalization ability even in small sample size.

Improving Harris hawks optimization algorithm for hyperparameters estimation and feature selection in v‐support vector regression based on opposition‐based learning

“…The original problem in v‐SVR leads to convex quadratic programming with inequality constraints as 44–50 where 0 ≤ v ≤ 1. It was proved that v is an upper bound on the fraction of margin errors and a lower bound of the fraction of support vectors by Schölkopf et al 5 Furthermore, by introducing v and adding an inequality constraint ε ≥ 0 in Equation , the value of can be automatically determined, and v is easier to determine than ε 5 …”

“…v‐SVR is a new category of promising nonlinear kernel that goal to find the best regression hyperplane with the smallest structural risk in high‐dimensional feature space 13 . In the SVR with ε‐insensitive loss function, the number of support vectors cannot be controlled 44 . In order to improve the solution speed of the SVR by controlling the number of support vector, training errors, and giving an estimate of the ε in the data, an improved version ν‐SVR was proposed by Schölkopf et al 5 …”

Performance Enhancement of Internal Combustion Engines through Vibration Control: State of the Art and Challenges

“…Engine mounting transmission ratio (EMTR), or transmissibility, which is defined as the ratio of the vibration generated at the engine to the one that is transmitted through the mounting system to the body of the vehicle, is a useful metric for expressing isolation efficiency of the engine mounts at a specific frequency or speed [14]. While many statistical tools have been developed to detect and analyze different physical and natural phenomena [15][16][17][18], a well-established measure in vibration diagnosis and analysis is the power content (root mean square or RMS) value of the measured vibration signal [19][20][21]. The transmission ratio of each mounting system at different engine speeds [14].…”

A study to determine the onset of catastrophic wear of a processing tool by statistical parameters of acoustic emission

“…Processing the set of new data helps classify the status of the cutting tool or the wear state of the cutting tool. The use of the principal component analysis in machining materials to assess the wear of a cutting tool has also been considered in a number of other studies -in the processing of Inconel 718 alloy [16], normalised steel [17], and Ti-6Al-4V alloy [18].…”