Gang Cheng

Fault diagnosis of transmission gears in the cutting drive of a coal-mining shearer is challenged by scarce field fault data and the production interruptions associated with drive failures. To alleviate performance degradation caused by blindly increasing network depth and to improve fault-feature extraction when real training data are limited, this study proposes a fault-diagnosis method that integrates a least-squares-loss and weight-normalization deep convolutional generative adversarial network (LW-DCGAN) with a convolutional-attention-enhanced 34-layer residual network (C-ResNet34). First, a particle-swarm-optimized symmetrized dot pattern (PS-SDP) transformation is used to convert vibration signals into image representations. The LW-DCGAN is then used to augment only the PS-SDP training images. Next, a convolutional attention mechanism is incorporated into the residual blocks of ResNet34 to construct C-ResNet34 and improve fault-feature extraction. The workflow was evaluated through image-transformation, multichannel-fusion, training-set-augmentation, and classifier-comparison experiments on a Drivetrain Diagnostics Simulator-based simulated shearer dataset and the public University of Connecticut gearbox dataset. Experimental results demonstrate the effectiveness of the proposed method.

Reliable remaining useful life (RUL) prediction of rolling bearings serves as the core support for the prognostics and health management (PHM) of industrial rotating equipment. Although data-driven methods have achieved remarkable breakthroughs in prediction accuracy, the existing approaches still encounter three critical limitations: the insufficient informational dimensions of individual features, the poor adaptability between degraded features and prediction tasks in mainstream feature fusion models, and the channel-wise and temporal redundancy in multi-dimensional features that masks the inherent degradation trends of bearings. To tackle these challenges, this paper presents a novel rolling bearing RUL prediction method based on multivariate time series mixers with cross-channel convolutional fusion (CCF) (MTS-mixers-CCF). The proposed method initially constructs an initial feature set from bearing vibration signals. Subsequently, it establishes a feature-task adaptability evaluation framework using an adaptive differential evolution based on comprehensive evaluation metrics (CEM) algorithm to reconstruct a high-adaptability feature ensemble. Finally, it attains accurate RUL estimation through the MTS-mixers-CCF model, which reduces feature redundancy by means of factorization mechanisms and enhances inter-feature correlation via CCF layers. Experiments conducted on the widely adopted PHM2012 and XJTU‐SY bearing datasets demonstrate that the MTS-mixers-CCF model outperforms traditional time series prediction methods and state of the art deep learning models, exhibiting significantly higher accuracy and stability for RUL prediction under ambiguous degradation trends. This research offers a robust and high-performance solution for rolling bearing RUL estimation, with promising application prospects in industrial PHM scenarios.

Early fault vibration signals from rolling bearings are typically nonlinear, non-stationary, and heavily obscured by background noise, which severely impedes the accurate extraction of fault features. To overcome the limitations of traditional stochastic resonance (SR)—specifically the small-parameter restriction for high-frequency signals and the subjectivity in parameter selection—this paper proposes an adaptive SR envelope spectroscopy method based on particle swarm optimization (PSO) and local mean decomposition (LMD). First, a variable-scale transformation is introduced to compress the high-frequency fault signals into the effective frequency band required by the adiabatic approximation theory. Second, utilizing the global search capability of PSO, the potential well parameters of the bistable system are adaptively optimized by maximizing the output signal-to-noise ratio (SNR), thereby achieving optimal matching between the nonlinear system and the input signal. Finally, the enhanced signal is decomposed by LMD, and the sensitive components are selected for envelope spectrum analysis to identify fault characteristics. Experimental validation using the Case Western Reserve University bearing dataset demonstrates that the proposed method effectively amplifies weak fault signals under strong noise conditions, exhibiting superior feature extraction accuracy and noise robustness compared to traditional methods.

This work addresses the challenges of inadequate data acquisition and the limited availability of labeled samples for shearer rocker arm bearing faults by developing a data augmentation methodology that synergistically incorporates the Genetic Algorithm-optimized Wavelet Transform Symmetrical Dot Pattern (GA-WT-SDP) with a Wasserstein Conditional Generative Adversarial Network (WCGAN). In the initial step, the Genetic Algorithm (GA) is employed to refine the mapping parameters of the Wavelet Transform Symmetrical Dot Pattern (WT-SDP), facilitating the transformation of raw vibration signals into advanced and discriminative graphical representations. Thereafter, the Wasserstein distance in conjunction with a gradient penalty mechanism is introduced through the WCGAN, thereby ensuring higher-quality generated samples and improved stability during model training. Experimental results validate that the proposed approach yields accelerated convergence and superior performance in sample generation. The augmented data significantly bolsters the generalization ability and predictive accuracy of fault diagnosis models trained on small datasets, with notable gains achieved in deep architectures (CNNs, LSTMs). The research substantiates that this technique helps overcome overfitting, enhances feature representation capacity, and ensures consistently high identification accuracy even in complex working environments.

Wind turbine drivetrains operate under variable aerodynamic loading, fluctuating speed, and harsh field environments, making condition monitoring difficult when fault samples are scarce and imbalanced. To address this challenge, this paper proposes a small‐sample data augmentation method that fuses variational mode decomposition–particle swarm optimization–symmetric dot pattern (VMD–PSO–SDP) feature optimization with a prior knowledge–guided Wasserstein conditional generative adversarial network (PK‐WCGAN). First, VMD is used to decompose vibration signals into informative modal components. In combination with an improved PSO algorithm for optimizing SDP parameters, high‐order fault feature maps are generated to enhance feature transferability. Then, real feature maps are used to pretrain the discriminator as prior knowledge, guiding the Wasserstein conditional generative adversarial network (WCGAN) to learn the target domain distribution. A gradient penalty mechanism is introduced to improve the quality of generated samples and training stability. Finally, generated samples are fused with real training samples to construct a hybrid dataset. Experimental results demonstrate that the generated data effectively reduce small‐sample distribution mismatch. When the target data samples are expanded from 50 to 1000, the average recognition accuracy of the five models increases from 54.6% to 94.0%, and the risk of overfitting is reduced by up to 26.5 percentage points under the 10‐sample setting. This method provides an effective solution for small‐sample fault diagnosis in the wind turbine drivetrain.

To address the difficulty in extracting early bearing fault signatures in the presence of strong background noise, a parameter-optimized feature mode decomposition (POFMD) method is proposed. First, a fitness function is constructed by combining dispersion entropy (DE) and the characteristic energy ratio. The Archimedes optimization algorithm (AOA) is then used to globally optimize the filter length L and the number of decomposed modes n in the feature mode decomposition (FMD) method, addressing the sensitivity of traditional FMD decomposition results to preset parameters. Second, the L and n corresponding to the maximum fitness value are selected as the preset FMD parameters for signal decomposition. Finally, envelope spectrum analysis is used to perform fault diagnosis. Through the analysis of signals from two public datasets (CWRU and XJTU-SY) and a bearing fault test bench, the proposed method exhibits excellent anti-noise capability and effective early weak fault diagnosis performance, providing a reliable solution for early bearing fault diagnosis.