Search

• The RUL prognosis framework combines the wavelet transform and the ConvLSTM. • Channel self-attention highlights weak but crucial features in wavelet coefficients. • Global attention improves the representation of the learned features in multiple wavelet coefficients of different channels. • Hybrid attention-based method boosts convlstm's performance in RUL prognosis. Wavelet transform, a time-frequency analysis method for evaluating non-stationary signals, can assist in representing equipment degradation over prolonged usage. However, a single wavelet basis function is challenging to apply to all periodic transient waveforms. As a result, this research suggests a hybrid attention-based multi-wavelet coefficient fusion method for evaluating the remaining useful life (RUL) of bearings. Firstly, a two-dimensional map is created by organizing the decomposed individual frequency bands after the approach employs several wavelets to get the original signal properties. Secondly, a hybrid attention-based ConvLSTM (HA-ConvLSTM) network is designed to weight wavelet coefficient channels adaptively. The learned features are used to evaluate RULs by a multi-layer perceptron. Finally, tests were run on the PHM2012 rolling bearing dataset to validate the proposed method. Overall, the suggested scheme outperforms previous comparable methods in the performance index. This approach optionally resolves the wavelet basis function matching issue for periodic transient waveforms. [ABSTRACT FROM AUTHOR]

• Tool structure parameters, process parameters, etc. are introduced to the model structure design. • Frequency and spatial components can be identified and weighted adaptively by designed attention modules. • The contribution of each attention module of the model is analyzed in detail through ablation experiments. • By visualizing and comparing the attention weights and features, the interpretability is explored. Real-time and accurate monitoring of tool wear status is critical for optimizing product quality and production costs. Deep learning has received much attention as an emerging data-driven technique in the field of tool wear monitoring because of its powerful feature extraction and nonlinear mapping capabilities. However, the model structure and features' poor interpretability limits the application of deep learning in practical machining. Herein, a frequency-spatial hybrid attention network is proposed, driven by tool structure and process parameters (FSHAN-SPD). The structure of FSHAN-SPD considers the recognition of critical frequencies and spatial positions simultaneously, and the receptive field of its spatial attention module can effectively distinguish the cutting stage of each tooth. Attention modules-weighted features are used for interpretability analysis. The effectiveness and flexibility of this method are verified in the PHM2010 dataset. A milling experiment illustrates a more detailed procedure. The experimental results are well interpretable in both frequency and spatial. [ABSTRACT FROM AUTHOR]