A Hybrid Attention-Based Paralleled Deep Learning model for tool wear prediction

作者：

Highlights：

• A parallel deep learning model is present to predict tool wear even under noisy case.

• Parallel structure is proposed to learn features without inner mutual interference.

• Block-attention units stresses sensitive component and improve model performance.

• Experiments are conducted to validate model performance and robustness thoroughly.

摘要

•A parallel deep learning model is present to predict tool wear even under noisy case.•Parallel structure is proposed to learn features without inner mutual interference.•Block-attention units stresses sensitive component and improve model performance.•Experiments are conducted to validate model performance and robustness thoroughly.

论文关键词：Attention mechanism,Convolution neural network,Deep learning,Recurrent neural network,Tool condition monitoring

论文评审过程：Received 6 February 2022, Revised 26 July 2022, Accepted 12 August 2022, Available online 18 August 2022, Version of Record 26 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118548