In intelligent fault diagnosis for rolling bearings, the variable operating conditions of rotating machinery lead to inconsistent distributions between training and testing data, inevitably decreasing fault diagnosis accuracy. To overcome this issue, we propose a bearing fault diagnosis method based on transfer learning with a dual-flow manifold residual network and improved capsule network (DMRCN). Firstly, we introduce a dual-flow manifold residual network structure consisting of continuous wide-kernel convolution, continuous narrow-kernel convolution, and manifold learning spatial information description blocks. This architecture is utilized to extract low-level features from the data. Then, we present a novel architecture of the capsule network with self-attention output. We incorporate a self-attention mechanism into the dynamic routing algorithm to transfer richer high-level features from the low-level capsule layer to the higher-level capsule layer. During training, we propose a loss function known as global-local distribution discrepancy loss to enhance the model’s generalization ability and accelerate model convergence. Finally, we validate the performance of DMRCN in cross-condition transfer fault diagnosis on both the Case Western Reserve University public dataset and a laboratory-built bearing experimental rig dataset. Experimental results demonstrate that DMRCN outperforms other commonly used deep transfer learning methods regarding fault diagnosis across different operating conditions.

To address the low accuracy of current one-dimensional signal recognition for coal-rock cutting vibration and the low efficiency of traditional static neural networks, this paper proposes an interpretable recognition method that combines Markov transition field (MTF) and channel-selective neural networks. Firstly, by decomposing and reconstructing the energy of wavelet packets and denoising the signal, it is transformed into MTF images with temporal correlation to improve signal processing efficiency. Then, a channel selective module is proposed to replace traditional convolutional layers, enhancing the model’s ability to extract data features. A spatial shift mechanism is improved to alleviate the problem of weight degradation and improve the model’s generalization capability. Experimental results on the self-built dataset show that MTF can effectively capture the dynamic characteristics of time series data compared to other signal transformation methods. ResNet + SelectConv achieved recognition accuracy of 97.375% without adding extra parameters and maintained good robustness in noise environments up to 6 dB, significantly outperforming traditional methods. This study improves the efficiency and accuracy of coal-rock cutting state recognition and provides new insights for signal processing in environments with limited underground computational resources.

Rolling bearings are a key component for ensuring the safe and smooth operation of rotating machinery and are very prone to failure. Therefore, intelligent fault diagnosis research on rolling bearings has become a crucial task in the field of mechanical fault diagnosis. This paper proposes research on the fault diagnosis of rolling bearings based on an adaptive nearest neighbor strategy and the discriminative fusion of multi-feature information using supervised manifold learning (AN-MFIDFS-Isomap). Firstly, an adaptive nearest neighbor strategy is proposed using the Euclidean distance and cosine similarity to optimize the selection of neighboring points. Secondly, three feature space transformation and feature information extraction methods are proposed, among which an innovative exponential linear kernel function is introduced to provide new feature information descriptions for the data, enhancing feature sensitivity. Finally, under the adaptive nearest neighbor strategy, a novel AN-MFIDFS-Isomap algorithm is proposed for rolling bearing fault diagnosis by fusing various feature information and classifiers through discriminative fusion with label information. The proposed AN-MFIDFS-Isomap algorithm is validated on the CWRU open dataset and our experimental dataset. The experiments show that the proposed method outperforms other traditional manifold learning methods in terms of data clustering and fault diagnosis.

Fault diagnosis of belt conveyors is crucial for coal mine production, but audio-based fault diagnosis in underground coal mines remains challenging due to the strong noise environment. To address this problem, a method for audio fault diagnosis of belt conveyors based on improved variational modal decomposition and improved adaptive noise reduction convolutional networks (IANCNs) in strong noise environments is proposed. Firstly, the improved beluga whale optimization is designed by introducing the non-linear balance factor and non-linear probability and combining them with the proposed cyclical shock factor to optimize the variational modal decomposition parameters to achieve noise reduction and signal reconstruction. Secondly, an IANCN is developed using an adaptive threshold activation function and an improved loss function to enhance noise robustness and fault diagnosis accuracy. Finally, the proposed method’s effectiveness is evaluated in low and strong noise environments, with experimental results demonstrating superior fault diagnosis performance. In low noise environments, the fault diagnosis accuracy is 98.61%, and in strong noise environments, it is 98.96%, outperforming existing fault diagnosis methods.