Jul 2026· Frontiers of Mechanical Engineering· 0 citations· 23 references
TL;DR
This smart fault diagnosis method based on the Time Convolution Network - Bidirectional Gated Recurrent Unit - Attention Model (TCN-BiGRU-Attention) can achieve high-precision and stable fault diagnosis for rolling bearings, providing an effective intelligent diagnosis solution for engineering applications.
Abstract
In complex working environments and noisy conditions, the vibration signals of bearings are highly non-stationary, and the fault characteristics are easily masked by noise, making it difficult to effectively identify the faults. Moreover, existing methods are relatively sensitive to parameter and working condition changes and lack diagnostic stability.
To address this issue, a smart fault diagnosis method based on the Time Convolution Network - Bidirectional Gated Recurrent Unit - Attention Model (TCN-BiGRU-Attention) was proposed. This method uses the Newton-Raphson optimization algorithm to optimize the parameters of the feature mode decomposition, extracts the feature modes, and combines them with the TCN-BiGRU-Attention deep temporal sequence model to achieve multi-scale feature modes and key temporal discrimination.
The experiments were conducted based on two public datasets - Case Western Reserve University and XJTU-SY, with each group of experiments repeated at least 10 times under the same initial conditions. At the same time, ablation experiments and performance comparison experiments with other advanced methods were carried out. The results show that in the fault identification task, the accuracy of this research method reached 96.32%, which is higher than 89.15% of one-dimensional convolutional neural networks, 91.45% of bidirectional long short-term memory neural networks, and 92.84% of time convolutional networks.
In conclusion, this method can achieve high-precision and stable fault diagnosis for rolling bearings, providing an effective intelligent diagnosis solution for engineering applications.
To tackle the persistent challenges of low bearing fault diagnosis accuracy—specifically the difficulty of extracting faint fault features under complex, variable operating conditions and strong background noise, as well as the tendency to lose deep temporal dependencies—this paper proposes a novel diagnosis method integrating FFT-VMD feature extraction with a Bi-TCN-Bi-GRU neural network. First, to overcome the heuristic parameter selection and frequency-band aliasing inherent in traditional Variational Mode Decomposition (VMD), the Fast Fourier Transform (FFT) is introduced. By extracting global spectral prior information, the FFT guides the VMD to perform adaptive decomposition and effective denoising of non-stationary vibration signals. Subsequently, a dual-path Bi-TCN-Bi-GRU diagnostic model is constructed. The dilated causal convolution mechanism of the Bidirectional Temporal Convolutional Network (Bi-TCN) is utilized to extract deep local spatial features, while the Bidirectional Gated Recurrent Unit (Bi-GRU) is integrated to deeply mine the forward and backward dynamic evolutionary dependencies within the sequence. Experimental results demonstrate that even under severe background noise with a signal-to-noise ratio (SNR) of -2 dB, the proposed method maintains an exceptional diagnostic accuracy of 98.21% on a self-built bearing dataset and 99.90% on the Southeast University (SEU) dataset. This study provides a highly robust and promising solution for enhancing bearing safety and predictive maintenance in complex industrial environments.
Bearing fault diagnosis is a key strategy to ensure the stability of mechanical systems, optimize maintenance plans and improve operational reliability. Vibration signals are complex time series with unique properties. Most of the current methods only consider the spatial characteristics of the signal, but do not take into account its temporal characteristics. In fact, vibration signals contain both spatial and temporal information, offering not only rich temporal dynamic details but also spatial structural insights that reflect fault characteristics. Therefore, in order to fully mine and integrate the spatiotemporal feature information in vibration signals to enhance the accuracy and robustness of intelligent diagnosis, we propose a bearing fault diagnosis method based on the fusion of spatial and temporal features. Firstly, an improved wide kernel deep convolutional neural network method was proposed. By using an adaptive channel module, the model was enabled to focus on the key features of the vibration signal and suppress the interference of irrelevant information. At the same time, the superimposed bidirectional long short-term memory layer and the gated recurrent unit effectively capture the long-term dependencies of the time series in the dataset, covering both past and future signal features. This method can effectively utilize temporal information and combine it with spatial features, significantly improving the accuracy and robustness of bearing diagnosis. To verify the proposed method, ablation and comparative experiments were conducted on two publicly available bearing datasets: the Case Western Reserve University dataset and the Guangdong University of Petrochemical Technology dataset. The experimental results show that the average accuracy rates of fault diagnosis of the WBLG network on the two sets of datasets have reached 99.7% and 96.5% respectively. Compared with the existing models, the maximum improvement rates are 3.73% and 12.4% respectively. It demonstrates its superior classification performance and generalization applicable to bearing fault diagnosis.
Qi Wang, Rui Huang, Yongda Cai et al.· Measurement science and tech...· 0 citations
To address the degradation of cross-condition diagnostic performance caused by feature-scale drift in rolling bearing vibration signals under variable operating conditions, this paper proposes a spectral-guided adaptive multi-scale convolutional neural network (SAMACNN). First, PSD sequences and time-frequency features are introduced as dual-stream inputs. While the time-frequency main branch extracts local information, the Spectral Transformer spectral bypass branch captures long-range dependencies in harmonic structures. Second, dynamic gating weights are generated for the multi-scale convolutional branches, enabling sample-conditioned multi-scale feature selection and fusion and alleviating scale mismatch caused by fixed receptive fields and static fusion. Finally, data collected from two bearing fault simulation test rigs are used to verify the effectiveness and superiority of the proposed algorithm. The experimental results show that the proposed SAMACNN method achieves average accuracies of 97.16% and 97.56% on the two datasets, respectively, outperforming the ablation variants and demonstrating strong robustness and generalization capability in complex variable-condition measurement environments.
DaXin Li, Wang Hong, Hai Xue et al.· Engineering Research Express· 0 citations
In this study, a hybrid deep learning architecture is proposed for robust vibration-based fault diagnosis in industrial machinery by jointly modeling time-domain, frequency-domain, and temporal dynamics. In the time domain, convolutional layers combined with an Enhanced Gated Attention (EGA) mechanism emphasize informative signal components while suppressing noise. Temporal evolution is modeled using Neural Ordinary Differential Equations (Neural ODEs), enabling smooth and stable continuous-time feature representations. In parallel, a Fourier Neural Operator (FNO) extracts frequency-domain characteristics, augmented with gated attention to focus on fault-related spectral patterns. Long Short-Term Memory (LSTM) layers capture long-range dependencies, while Squeeze-and-Excitation (SE) blocks adaptively recalibrate channel-wise feature responses. A multi-scale attention-based fusion module integrates domain-specific representations and auxiliary features to enhance discrimination under varying operating conditions. The proposed model is evaluated on the SUBFv1.0 dataset through extensive ablation studies and experiments under multiple noise levels, achieving 98.41% accuracy in noise-free conditions and maintaining performance above 91% even at 5 dB SNR. Unlike existing multi-path approaches that combine heterogeneous features in a loosely coupled or discrete manner, the proposed architecture uniquely integrates multi-domain feature learning with bidirectional attention mechanisms and continuous-time temporal dynamics, enabling coherent cross-domain interaction and robust fault characterization.
Canan Taştimur· Information Technology and C...· 0 citations
A dual-channel attention and feature selection-based fault diagnosis method for rolling bearings is proposed, and a corresponding digital twin-based interaction system is further developed.
Erpeng Wang, Zhaoze Sun, Jian Wang et al.· Engineering Research Express· 0 citations
A rolling bearing fault diagnosis method based on multi-scale depthwise separable convolution (MDSC) and a convolutional neural network–Transformer hybrid model (CNN-Transformer) is proposed to address the non-stationarity of fault signals and the difficulty of jointly capturing local and global features. First, continuous wavelet transform (CWT) converts one-dimensional vibration signals into two-dimensional time-frequency images to enhance fault representation. Then, multi-scale convolution (MSC) and depthwise separable convolution (DSC) are introduced to extract local impulsive features and fault patterns at different scales with fewer parameters. A CNN-Transformer architecture is further developed, where convolutional neural network (CNN) captures local details and Transformer models global dependencies. In addition, pretraining-finetuning, data augmentation, label smoothing, and normal sample optimization are adopted to improve training stability and diagnostic performance. Experimental results show accuracies of 98.80% on the Xi’an Jiaotong University bearing dataset (XJTU-SY) and 100.00% on the Case Western Reserve University bearing dataset (CWRU), demonstrating strong discriminative ability, stability, and robustness.
Shuai Yang, Yanchao Chen, Yang Yu· Engineering Research Express· 0 citations