Tool wear state recognition based on feature selection method with whitening variational mode decomposition

Wei, Xudong; Liu, Xianli; Yue, Caixu; Wang, Lihui; Liang, Steven Y.; Qin, Yiyuan

doi:10.1016/j.rcim.2022.102344

Cited by 21 publications

(4 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The IWOA algorithm's population size was set to 50, and it was programmed to iterate a maximum of 20 times. The mode count was [1,11] K  , and the penalization coefficient was [500,3000]   . Using the signal of the cutting force in the x-direction during the normal wear phase of the milling cutter as a case study, and at the same time, to mitigate the influence of the tool wear condition at the beginning and end of the cutting process, this study employed a sample dataset consisting of 4096 data points for each cutting cycle, ranging from 60,001 to 64,096.…”

Section: Signal Analysis Of Milling Cutter Wear Cutting Forcementioning

confidence: 99%

“…There is presently no obvious foundation for selecting scale factors. As a result, in order to discover the appropriate scaling factor, this study examined the change in MPE with scale factor s ∈ [1,18], as shown in Figure 10. lowest, resulting in an inability to accurately depict the milling cutter's wear process.…”

Section: Mpe-based Milling Cutter Wear State Feature Extractionmentioning

confidence: 99%

“…As a result, the real-time and exact monitoring of tool wear during the machining process of thin-walled components is critical for improving machining precision and productivity. Current methodologies for the monitoring cutting tool status are divided into two categories: direct inspection techniques and indirect inspection techniques [1]. Currently, the most common direct monitoring approach is machine vision, but the technology is costly, and the picture is readily altered by onsite processing circumstances [2].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Tool Wear State Identification Based on the IWOA-VMD Feature Selection Method

Shui,

Rong,

Dan

et al. 2024

Machines

View full text Add to dashboard Cite

Complex, thin-walled components are the most important load-bearing structures in aircraft equipment. Monitoring the wear status of milling cutters is critical for enhancing the precision and efficiency of thin-walled item machining. The cutting force signals of milling cutters are non-stationary and non-linear, making it difficult to detect wear stages. In response to this issue, a system for monitoring milling cutter wear has been presented, which is based on parameterized Variational Mode Decomposition (VMD) Multiscale Permutation Entropy. Initially, an updated whale optimization technique is used, with the joint correlation coefficient serving as the fitness value for determining the VMD parameters. The improved VMD technique is then used to break down the original signal into a series of intrinsic mode functions, and the Multiscale Permutation Entropy of each effective mode is determined to generate a feature vector. Finally, a 1D Convolutional Neural Network (1D CNN) is employed as the input model for state monitoring using the feature vector. The experimental findings show that the suggested technique can efficiently extract characteristics indicating the wear condition of milling cutters, allowing for the precise monitoring of milling cutter wear states. The recognition rate is as high as 98.4375%, which is superior to those of comparable approaches.

show abstract

Section: Signal Analysis Of Milling Cutter Wear Cutting Forcementioning

confidence: 99%

Section: Mpe-based Milling Cutter Wear State Feature Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tool Wear State Identification Based on the IWOA-VMD Feature Selection Method

Shui,

Rong,

Dan

et al. 2024

Machines

View full text Add to dashboard Cite

show abstract

“…Duan et al [12] converted cutting signals into time-frequency maps by short-time Fourier transform, then feature extraction by PCANet and GA-SPP, and finally tool wear prediction by SVM. Wei et al [13] screened the sensitive low-dimensional features of force signals by whitening variational mode decomposition (WVMD) and Joint information entropy (JIE), and optimal-path forest (OPF) was used as a classifier to realize the tool wear state recognition. Chan et al [14] achieved tool wear state monitoring by extracting overall and local features of the signal through LSTM.…”

Section: Introductionmentioning

confidence: 99%

Tool Wear State Recognition Based on One-Dimensional Convolutional Channel Attention

Xue,

Li,

Chen

et al. 2023

Micromachines

View full text Add to dashboard Cite

Tool wear state recognition is an important part of tool condition monitoring (TCM). Online tool wear monitoring can avoid wasteful early tool changes and degraded workpiece quality due to later tool changes. This study incorporated an attention mechanism implemented by one-dimensional convolution in a convolutional neural network for improving the performance of the tool wear recognition model (1DCCA-CNN). The raw multichannel cutting signals were first preprocessed and three time-domain features were extracted to form a new time-domain sequence. CNN was used for deep feature extraction of temporal sequences. A novel 1DCNN-based channel attention mechanism was proposed to weigh the channel dimensions of deep features to enhance important feature channels and capture key features. Compared with the traditional squeeze excitation attention mechanism, 1DCNN can enhance the information interaction between channels. The performance of the model was validated on the PHM2010 public cutting dataset. The excellent performance of the proposed 1DCCA-CNN was verified by the improvement of 4% and 5% compared to the highest level of existing research results on T1 and T3 datasets, respectively.

show abstract