Intelligent fault diagnosis for rotary machinery often suffers performance degradation under domain shifts between training and deployment environments. To address this, we propose a robust fault diagnosis framework incorporating three key components: (1) an order-frequency-based preprocessing method to normalize rotational variations, (2) a U-Net variational autoencoder (U-NetVAE) to enhance adaptation through reconstruction learning, and (3) a test-time training (TTT) strategy enabling unsupervised target domain adaptation without access to source data. Since existing works rarely evaluate under true domain shift conditions, we first construct a unified cross-domain benchmark by integrating four public datasets with consistent class and sensor settings. The experimental results show that our method outperforms conventional machine learning and deep learning models in both F1-score and recall across domains. Notably, our approach maintains an F1-score of 0.47 and recall of 0.51 in the target domain, outperforming others under identical conditions. Ablation studies further confirm the contribution of each component to adaptation performance. This study highlights the effectiveness of combining mechanical priors, self-supervised learning, and lightweight adaptation strategies for robust fault diagnosis in the practical domain.

Prognostics and health management of general rotating machinery have been studied over time to improve system stability. Recently, the excellent abnormal diagnosis performance of artificial intelligence (AI) was demonstrated, and therefore, AI-based intelligent diagnosis is now being implemented in these systems. AI models are trained using large volumes of data. Therefore, we propose a transformer-based generative adversarial network (GAN) model with a multi-resolution short-time Fourier transform (multi-STFT) loss function to augment the vibration data of rotating machinery to facilitate the successful learning of deep learning models. We constructed a model with a conditional GAN structure, which is transformer based, for learning the feature points of vibration data in the time-series domain. In addition, we applied the multi-STFT loss function to capture the frequency features of the vibration data. The generated data, which adequately captured the frequency features, were used to augment the training data to improve the performance of a deep learning classifier. Furthermore, by visualizing the generated vibration data and comparing the visualizations to those of the vibration data obtained from real machinery, we demonstrated that the generated data were indistinguishable from the actual data.

In the domain of prognosis and health management (PHM) for rotating machinery, the criticality of ensuring equipment reliability cannot be overstated. With developments in artificial intelligence (AI) and deep learning, there have been numerous attempts to use those methodologies in PHM. However, there are challenges to applying them in practice because they require huge amounts of data. This study explores a novel approach to augment vibration data—a primary component in traditional PHM methodologies—using a specialized generative model. Recognizing the limitations of deep learning models, which often fail to capture the intrinsic physical characteristics vital for vibration analysis, we introduce the bivariate vibration generative adversarial networks (BiVi-GAN) model. BiVi-GAN incorporates elements of a physics-informed neural network (PINN), emphasizing the specific vibration characteristics of rotating machinery. We integrate two types of physical information into our model: order analysis and cross-wavelet transform, which are crucial for dissecting the vibration characteristics of such machinery. Experimental findings show the effectiveness of our proposed model. With the incorporation of physics information (PI) input and PI loss, the BiVi-GAN showed a 70% performance improvement in terms of JS divergence compared with the baseline biwavelet-GAN model. This study maintains the potential and efficacy of complementary domain-specific insights with data-driven AI models for more robust and accurate outcomes in PHM.

The degradation of clamping force in the core support barrel, which forms the internal structure of a nuclear power plant, has the potential to significantly impact the plant’s safety and reliability. Previous studies have concentrated on the detection of clamping force degradation but have been constrained in their ability to identify the precise size and position. This study proposes a novel methodology for diagnosing the size and position of clamping force degradation in core support barrels, combining deep-learning techniques and dynamic time warping (DTW) algorithms. DTW is applied to the magnitude data of the ex-core neutron noise signal obtained in the frequency domain, thereby enabling the effective learning of changes in sensor data values. Moreover, autoencoder-based (AE-based) representation learning is utilized to extract features of the data, preventing overfitting and thus enhancing the robustness of the model. The experiment results demonstrate that the size and position of clamping force degradation can be accurately predicted. It is expected that this research will contribute to enhancing the precision and efficiency of internal structure monitoring in nuclear power plants.

Origin–destination (OD) matrix forecasting is essential for urban railway operations because it enables simultaneous understanding of the direction and magnitude of passenger flows. However, OD matrices in large-scale subway networks are difficult to predict owing to their high dimensionality and sparsity, and existing approaches often rely on station-level predictions or complex structural designs. This study addresses the offline full-matrix OD forecasting problem, where complete historical OD sequences are available at prediction time, and proposes Metro-GATF, a spatiotemporal forecasting framework that jointly models railway topology and dynamic OD interactions. The model employs a GATv2-based spatial encoder to learn static inter-station relationships and encodes time-varying interactions using sparse OD graphs. A non-autoregressive transformer decoder generates future multi-step node representations in parallel, whereas origin–destination factorization and sparsity-aware gating are used to reconstruct the full OD matrix. Experiments on minute-level AFC-based OD data from a 637-station metropolitan subway network demonstrated that Metro-GATF achieved the lowest sMAPE among the compared full-matrix models. These results indicate that the proposed framework effectively captures complex spatiotemporal OD patterns and offers a practical end-to-end framework for forecasting urban railway demand.