Neural Network and Large Model-Driven Fault Diagnosis and Intelligent Operation and Maintenance for Rotating Machinery

Dear Colleagues,

Being at the "industrial heart" of key fields such as energy and power, aerospace, automotive manufacturing, petrochemical engineering, and new energy, rotating machinery's operational reliability is directly related to industrial safety, production efficiency, and economic benefits. However, under high-speed, heavy-load, and variable working conditions, as well as complex service environments, rotating machinery is prone to various faults, such as bearing wear, rotor imbalance, and gear damage. If not detected and handled in a timely manner, these faults may lead to equipment shutdown or significant accidents.

Traditional fault diagnosis, operation, and maintenance methods rely heavily on manual feature extraction and expert experience, which have limitations such as low efficiency, poor generalization ability, and difficulty in adapting to complex scenarios involving hydraulic turbine units and offshore wind turbines. With the rapid development of artificial intelligence technology, neural networks and large models have shown remarkable advantages related to processing massive unstructured data, mining hidden fault patterns, and realizing end-to-end intelligent decision-making, from adaptive feature extraction by models such as Convolutional Neural Networks (CNNs) and Transformers to LLM-based zero-shot fault diagnosis frameworks. These technologies effectively address the pain points of traditional methods, such as their reliance on prior knowledge and sensitivity to environmental interference, and have become the subject of core research directions in the field of rotating machinery operation and maintenance.

To promote the in-depth integration of neural networks, large models, and fault diagnosis/intelligent operation and maintenance of rotating machinery—especially focusing on technological breakthroughs in typical equipment such as hydraulic turbine units and offshore wind turbines—and to showcase the latest research achievements and practical applications in this field, this Special Issue of Computation is being launched. We invite researchers, engineers, and practitioners from academia and industry to contribute high-quality original research papers, review articles, and technical notes.

The scope of this Special Issue includes, but is not limited to, the following topics:

• Design and optimization of neural network models for rotating machinery, such as CNN, RNN/LSTM, Transformer, and Graph Neural Networks;
• Application of large language models (LLMs) and foundation models in intelligent operation and maintenance of rotating machinery, including zero-shot fault diagnosis, fault reasoning, maintenance knowledge graph construction, and operation and maintenance text analysis;
• Data-driven condition monitoring and fault prediction for rotating machinery, such as multi-sensor data fusion, non-Gaussian signal processing, data augmentation technologies, and anomaly detection;
• Development of intelligent operation and maintenance systems for rotating machinery, including digital twin modelling and simulation, mixed-reality technology application, edge–end diagnosis system deployment, and lightweight model design;
• Fault diagnosis technologies in complex scenarios, such as variable working condition adaptation, cross-domain transfer learning, few-shot learning, and reliability optimization in extreme environments;
• Industrial application cases, including fault diagnosis systems for hydraulic turbine units, intelligent monitoring platforms for offshore wind turbines, and engineering practice of predictive maintenance for rotating machinery;
• Technical challenges and solutions, such as enhanced model interpretability, real-time optimization, data quality improvement, and multi-source heterogeneous data fusion.

We believe that this Special Issue will provide an important platform for academic exchange and technological cooperation in related fields to promote the industrial application of innovative achievements in key equipment such as hydraulic turbine units and offshore wind turbines. All submissions will undergo a rigorous single-blind peer-review process in accordance with MDPI's standards. We look forward to receiving your contributions.

Dr. Xueyi Li
Dr. Tianyang Wang
Dr. Haochen Qi
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Computation is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• rotating machinery
• fault diagnosis
• remaining useful life
• neural networks
• large models
• intelligent operation and maintenance
• predictive maintenance
• hydraulic turbine units
• offshore wind turbines
• digital twin
• fault prediction and health management (PHM)