In response to the challenge of small and imbalanced Datasets, where the total Sample size is limited and healthy Samples significantly outweigh faulty ones, we propose a diagnostic framework designed to tackle Class imbalance, denoted as the Dual-Stream Adaptive Deep Residual Shrinkage Vision Transformer with Interclass–Intraclass Rebalancing Loss (DSADRSViT-IIRL). Firstly, to address the issue of limited Sample quantity, we incorporated the Dual-Stream Adaptive Deep Residual Shrinkage Block (DSA-DRSB) into the Vision Transformer (ViT) architecture, creating a DSA-DRSB that adaptively removes redundant signal information based on the input data characteristics. This enhancement enables the model to focus on the Global receptive field while capturing crucial local fault discrimination features from the extremely limited Samples. Furthermore, to tackle the problem of a significant Class imbalance in long-tailed Datasets, we designed an Interclass–Intraclass Rebalancing Loss (IIRL), which decouples the contributions of the Intraclass and Interclass Samples during training, thus promoting the stable convergence of the model. Finally, we conducted experiments on the Laboratory and CWRU bearing Datasets, validating the superiority of the DSADRSViT-IIRL algorithm in handling Class imbalance within mixed-load Datasets.

Rolling bearings play a crucial role in industrial equipment, and their failure is highly likely to cause a series of serious consequences. Traditional deep learning-based bearing fault diagnosis algorithms rely on large amounts of training data; training and inference processes consume significant computational resources. Thus, developing a lightweight and suitable fault diagnosis algorithm for small samples is particularly crucial. In this paper, we propose a bearing fault diagnosis algorithm based on probabilistic decoupling knowledge distillation and meta-learning (MIX-MPDKD). This algorithm is lightweight and deployable, performing well in small sample scenarios and effectively solving the deployment problem of large networks in resource-constrained environments. Firstly, our model utilizes the Model-Agnostic Meta-Learning algorithm to initialize the parameters of the teacher model and conduct efficient training. Subsequently, by employing the proposed probability-based decoupled knowledge distillation approach, the outstanding performance of the teacher model was imparted to the student model, enabling the student model to converge rapidly in the context of a small sample size. Finally, the Paderborn University dataset was used for meta-training, while the bearing dataset from Case Western Reserve University, along with our laboratory dataset, was used to validate the results. The experimental results demonstrate that the algorithm achieved satisfactory accuracy performance.

In light of the inadequacies of traditional P-wave arrival picking algorithms using long and short windows, which exhibit poor anti-noise ability and do not meet the requirements for efficient and accurate P-wave arrival picking, this study proposes a CGAS deep learning algorithm for P-wave arrival picking in mine micro-seismic events. The algorithm constructs the depth characteristics of microseismic waveform through deep learning training and converts the problem of picking up the microseismic waveform into a classification problem by employing data set segmentation and classification. This allows the algorithm to pick up the characteristic information of microseismic waveform and improve the accuracy of P-wave picking. The algorithm was applied to pick up the three-component micro-seismic waveforms obtained from the microseismic monitoring system in a mining area in Liaoning from 2019 to 2020. Through ablation experiments, it was found that adding each structure to the pickup model improved the relevant indexes to some extent. When all network structures were used, the accuracy of the P-wave pick-up model reached 98.61%, ensuring the accuracy of P-wave pickup upon arrival and demonstrating the effectiveness of each layer structure in the classification of P-wave arrival. Compared to other P-wave arrival picking algorithms such as STA/LTA, U-Net++, Dpick, PhaseNet and EQTransformer, the algorithm proposed in this paper exhibited high precision in P-wave arrival time picking, providing a new research idea and technical means for efficiently determining the P-wave arrival of events.

Deep learning-related technologies have achieved remarkable success in the field of intelligent fault diagnosis. Nevertheless, the traditional intelligent diagnosis methods are often based on the premise of sufficient annotation signals and balanced distribution of classes, and the model structure is so complex that it requires huge computational resources. To this end, a lightweight class imbalanced diagnosis framework based on a depthwise separable Laplace-wavelet convolution network with variable-asymmetric focal loss (DSLWCN-VAFL) is established. Firstly, a branch with few parameters for time-frequency feature extraction is designed by integrating wavelet and depthwise separable convolution. It is combined with the branch of regular convolution that fully learns time-domain features to jointly capture abundant discriminative features from limited samples. Subsequently, a new asymmetric soft-threshold loss, VAFL, is designed, which reasonably rebalances the contributions of distinct samples during the model training. Finally, experiments are conducted on the data of bearing and gearbox, which demonstrate the superiority of the DSLWCN-VAFL algorithm and its lightweight diagnostic framework in handling class imbalanced data.

The strip exit thickness is an important indicator for measuring the production quality of steel rolling products. Controlling its accuracy is crucial in producing high-quality rolling products in the steel industry. This study proposed an improved marine predators algorithm (IMPA) optimized bidirectional gated recurrent unit (BiGRU) for predicting the strip exit thickness. Firstly, an adaptive weight factor was introduced into the marine predators algorithm (MPA) to overcome the drawbacks of slow early convergence speed and limited search range of the algorithm, addressing the issue of local optimization. Secondly, an average fitness strategy was developed to enhance the quality of the algorithm population, improve optimization accuracy, and strengthen algorithm stability. In addition, IMPA was used to optimize parameters, reducing the impact of hyper-parameters such as the number of hidden layer neurons, learning rate, and batch size on the prediction accuracy of BiGRU, thereby establishing IMPA-BiGRU model. The experiments were conducted on an actual rolling process dataset. The experimental results showed that IMPA outperformed traditional MPA, particle swarm optimization (PSO), grey wolf optimization (GWO), whale optimization algorithm (WOA), and recent variants of MPA. The IMPA-BiGRU also demonstrated excellent prediction capabilities for strip exit thickness, meeting the practical production requirements for high-quality rolling products.