Special Session-11

Special Session Organizer(s):

Prof. Lei Hou, Harbin Institute of Technology, China

Prof. Haiming Yi, Harbin Institute of Technology, China

Prof. Zeyuan Chang, Harbin Institute of Technology, China

Assoc. Prof. Nasser A. Saeed, Menoufia University, Egypt

Introduction and Topics:

Rotating machinery is a core component in aero-engines, gas turbines, wind turbines, rail transit, and advanced manufacturing systems. Its health condition directly affects the safety, reliability, and efficiency of critical industrial equipment. However, fault diagnosis of rotating machinery still faces major challenges in real-world applications, including limited fault samples, imbalanced data, variable operating conditions, strong background noise, multimodal heterogeneity, and the continuous evolution of machine states over long-term service.

Recent advances in generative artificial intelligence (AI) provide new opportunities for intelligent fault diagnosis and prognostics. Generative models, such as generative adversarial networks, variational autoencoders, diffusion models, large language models, and multimodal foundation models, have shown strong potential in data augmentation, representation learning, missing-data completion, cross-domain transfer, uncertainty-aware reasoning, and human–AI collaborative decision-making. These developments are opening a promising research direction for more robust, adaptive, and scalable diagnosis methods for rotating machinery.

This special session aims to bring together researchers and practitioners from academia and industry to present the latest advances in generative AI-enabled methods for rotating machinery fault diagnosis, condition monitoring, and intelligent maintenance. Contributions are encouraged on both theoretical developments and engineering applications.

Topics of interest include, but are not limited to:

• Generative AI for rotating machinery fault diagnosis and health monitoring
• Diffusion models, GANs, VAEs, and foundation models for vibration signal generation and enhancement
• Synthetic data generation for small-sample, imbalanced, and rare-fault scenarios
• Multimodal generative learning with vibration, acoustic, image, text, and operational data
• Self-supervised, semi-supervised, and few-shot diagnosis using generative models
• Domain adaptation, transfer learning, and cross-machine/cross-condition generalization
• Generative AI for anomaly detection, unknown fault discovery, and open-set diagnosis
• Physics-informed and knowledge-guided generative diagnosis methods
• Digital twins and generative AI for intelligent operation and maintenance
• Interpretability, uncertainty quantification, and trustworthy AI in diagnosis systems
• Edge deployment and real-time applications of generative diagnostic models
• Industrial case studies and benchmark datasets for intelligent rotating machinery diagnosis