In recent years, deep learning- based methods have attracted much attention and achieved remarkable results for intelligent fault diagnosis of rotating machinery. However, in many actual industrial scenes, there are few labeled and large unlabeled samples that can be collected for learning due to the cost of experiments, which affect seriously the performance of supervised learning methods. A novel semi-supervised fault diagnosis method based on multiscale permutation entropy (MPE) enhanced contrastive learning (CL) is proposed. Firstly, each channel signal of labeled and unlabeled samples collected in fault diagnosis experiments are expanded into two-dimensional matrices by minimizing MPE and encoded into gray Euclidean distance plots (EPs). Secondly, these EPs of different channels of each sample are aggregated and spliced into a color EP. CL model is employed as a pre-training model to learn deep features of color EPs of all labeled and unlabeled samples. Finally, the few labeled samples are inputted into ResNet18 to obtain good classification performance with the learned feature weights by the CL pre-training model. The applications of gearbox and milling tool fault diagnosis experiments show the classification accuracy of the proposed MPE-CL method outperforms significantly other five benchmark methods, which are above 95.4% and 96.0% when the sample size in the labeled training dataset is 50 per class, respectively.

Accurate tool condition monitoring (TCM) is important for the development and upgrading of the manufacturing industry. Recently, machine-learning (ML) models have been widely used in the field of TCM with many favorable results. Nevertheless, in the actual industrial scenario, only a few samples are available for model training due to the cost of experiments, which significantly affects the performance of ML models. A time-series dimension expansion and transfer learning (TL) method is developed to boost the performance of TCM for small samples. First, a time-frequency Markov transition field (TFMTF) is proposed to encode the cutting force signal in the cutting process to two-dimensional images. Then, a modified TL network is established to learn and classify tool conditions under small samples. The performance of the proposed TFMTF-TL method is demonstrated by the benchmark PHM 2010 TCM dataset. The results show the proposed method effectively obtains superior classification accuracies for small samples and outperforms other four benchmark methods.

Transfer learning (TL) has been successfully implemented in tool condition monitoring (TCM) to address the lack of labeled data in real industrial scenarios. In current TL models, the domain offset in the joint distribution of input feature and output label still exists after the feature...

Accurate prediction of the State of Health (SOH) of lithium batteries is essential for ensuring safe operation, prolonging battery service life and optimizing energy management. Battery degradation is characterized by strong nonlinearity, multi-factor coupled interference and data redundancy. Furthermore, prevailing hybrid prediction models are constrained by several inherent limitations: insufficient feature extraction, limited optimization capacity of conventional algorithms, and oversimplified modular integration. To tackle these challenges, this paper proposes an SOH prediction method based on Variational Mode Decomposition (VMD) and the Transformer-Bidirectional Long Short-Term Memory (BiLSTM) framework optimized by the Multi-Strategy Adaptive Beaver Behavior Optimizer (MSA-BBO). First, multi-dimensional health features are extracted from raw data. Second, the VMD algorithm is employed to decompose and reconstruct multi-dimensional health features (such as charging voltage and current) in the training set, thereby suppressing noise and extracting critical aging characteristics. The obtained processing scheme is then directly applied to the test set. Subsequently, a Transformer-BiLSTM hybrid framework is established, where the Transformer captures long-range temporal dependencies and the BiLSTM extracts bidirectional correlation features of data, thereby synergistically overcoming the limitations of single-structure models. To address the deficiencies of the original Beaver Behavior Optimizer (BBO), including insufficient utilization of elite information, convergence oscillation, and non-targeted perturbation, the algorithm is improved in four aspects: population evolution, exploration-exploitation balance, local optimization, and global escape. The developed MSA-BBO is employed for adaptive hyperparameter optimization. In contrast to existing dual-module hybrid models, the proposed method establishes a three-stage collaborative framework that integrates feature processing, feature mining, and hyperparameter optimization. Finally, comparative experiments with five other models are conducted to objectively evaluate the prediction accuracy, stability, and generalization ability of the proposed model. The experimental results demonstrate that the proposed model achieves the optimal overall performance on five test sets, with all coefficients of determination (R²) above 0.94 and all Mean Absolute Percentage Error (MAPE) values below 0.04. This research can provide reliable technical support and theoretical reference for the full-life cycle health management of lithium batteries.