Fault diagnosis of rotating machinery using novel self-attention mechanism TCN with soft thresholding method
Li Ding,
Qing Li
Abstract:Rotating machinery (e.g., rolling bearing and gearbox) are usually operated in high-risk and vulnerable environments such as time-varying loads and poor lubrication. Timely assessment of the operational status for rotating machinery is crucial to prevent damage caused by potential failure and shutdown, which significantly enhances the reliability of mechanical systems, prolongs the service life of critical components in rotating machinery, and minimizes unnecessary maintenance costs. To this regard, in this pa… Show more
“…Unfortunately, noise interference is often ignored in these approaches. Many studies on noise-robust fault diagnosis have shown that using thresholding operations as a network layer for feature extraction from convolutional layers is a viable solution [27]. However, each soft and hard thresholding has limitations, so it is necessary to balance them without adding too much model complexity.…”
Deep learning-based methods have shown promising results in fault diagnosis, but research on interpretability and noise robustness still needs to be done. A multi-channel wide-kernel wavelet convolutional neural network is proposed to address these issues. Firstly, a first layer of multi-channel wide-kernel convolution is designed to fuse different weight information and suppress high-frequency noise. Secondly, a discrete wavelet transform block is designed to retain the low-frequency components of the discrete wavelet transform for signal denoising and feature dimension reduction. At the same time, Improved Balance Dynamic Adaptive Threshold is used to enhance the robustness of the model’s noise and the sparsity of features, making the model easier to optimize. Lastly, a power spectrum and normalized class activation mapping are designed to validate the post-hoc explanations of the model. The effectiveness and reliability of the Multi-Channel Wide Kernel Wavelet Convolutional Neural Network are verified through two gearbox datasets.
“…Unfortunately, noise interference is often ignored in these approaches. Many studies on noise-robust fault diagnosis have shown that using thresholding operations as a network layer for feature extraction from convolutional layers is a viable solution [27]. However, each soft and hard thresholding has limitations, so it is necessary to balance them without adding too much model complexity.…”
Deep learning-based methods have shown promising results in fault diagnosis, but research on interpretability and noise robustness still needs to be done. A multi-channel wide-kernel wavelet convolutional neural network is proposed to address these issues. Firstly, a first layer of multi-channel wide-kernel convolution is designed to fuse different weight information and suppress high-frequency noise. Secondly, a discrete wavelet transform block is designed to retain the low-frequency components of the discrete wavelet transform for signal denoising and feature dimension reduction. At the same time, Improved Balance Dynamic Adaptive Threshold is used to enhance the robustness of the model’s noise and the sparsity of features, making the model easier to optimize. Lastly, a power spectrum and normalized class activation mapping are designed to validate the post-hoc explanations of the model. The effectiveness and reliability of the Multi-Channel Wide Kernel Wavelet Convolutional Neural Network are verified through two gearbox datasets.
Damage to the composite propeller blades could lead to rotational imbalance, which seriously affects the operational safety of unmanned aerial vehicles (UAVs), therefore, a novel method combining the Teager energy operator and bidirectional temporal convolutional network is proposed for detecting, localizing, and quantifying the damage-related imbalance in the blades. A flexible sensing system that contains MEMS accelerometers, signal conditioning, and wireless transmission is integrated with the composite propeller for in-situ signal acquisition of the propeller blades. Teager energy operator (TEO) is applied to demodulate and enhance the pulse compositions in vibration signals and singular value decomposition (SVD) is employed to suppress random noise, resulting in denoised Teager energy spectrums for model input. Temporal convolutional network (TCN) has been widely used in sequence signal modeling because the causal dilated convolution could learn the context information of sequence signals while maintaining the advantages of parallel computing. To fully extract the signal features, bidirectional temporal convolutional network (BiTCN) models are established to learn both the forward and backward signal features. Experimental verification results show that the proposed method detects the existence of imbalance with 100% accuracy, and the accuracies of localization and quantization are 99.65% and 98.61%, respectively, which are much higher than those of the models with the original signal as input. In addition, compared with the other four different algorithms, BiTCN is superior in terms of convergence speed and prediction accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.