An aircraft engine remaining useful life prediction method based on predictive vector angle minimization and feature fusion gate improved transformer model

Journal of Manufacturing Systems, Год журнала: 2024, Номер 76, С. 567 - 584

Опубликована: Сен. 1, 2024

Язык: Английский

---

A novel method for predicting the remaining useful life of MOSFETs based on a linear multi-fractional Lévy stable motion driven by a GRU similarity transfer network

Reliability Engineering & System Safety, Год журнала: 2025, Номер unknown, С. 110818 - 110818

Опубликована: Янв. 1, 2025

Язык: Английский

---

Prediction of Remaining Useful Life of Aero-engines Based on CNN-LSTM-Attention

International Journal of Computational Intelligence Systems, Год журнала: 2024, Номер 17(1)

Опубликована: Сен. 3, 2024

Abstract Accurately predicting the remaining useful life (RUL) of aircraft engines is crucial for maintaining financial stability and aviation safety. To further enhance prediction accuracy engine RUL, a deep learning-based RUL method proposed. This possesses potential to strengthen recognition data features, thereby improving model. First, input features are normalized CMAPSS (Commercial Modular Aero-Propulsion System Simulation) dataset utilized calculate engines. After extracting attributes from using convolutional neural network (CNN), extracted into long short-term memory (LSTM) model, with addition attention mechanisms predict Finally, proposed model evaluated compared through ablation studies comparative experiments. The results indicate that CNN-LSTM-Attention exhibits superior performance datasets FD001, FD002, FD003, FD004, RMSEs 15.977, 14.452, 13.907, 16.637, respectively. Compared CNN, LSTM, CNN-LSTM models, demonstrates better across datasets. In comparison other this achieves highest on dataset, showcasing strong reliability accuracy.

Язык: Английский

---

Power Transformer Prognostics and Health Management Using Machine Learning: A Review and Future Directions

Machines, Год журнала: 2025, Номер 13(2), С. 125 - 125

Опубликована: Фев. 7, 2025

Power transformers (PTs) play a vital role in the electrical power system. Assessing their health to predict remaining useful life is essential optimise maintenance. Scheduling right maintenance for equipment at time ultimate goal of any system utility. Optimal has number benefits: human and social, by limiting sudden service interruptions, economic, due direct indirect costs unscheduled downtime. PT now produces large amounts easily accessible data increasing use IoT, sensors, connectivity between physical assets. As result, transformer prognostics management (PT-PHM) methods are increasingly moving towards artificial intelligence (AI) techniques, with several hundreds scientific papers published on topic PT-PHM using AI techniques. On other hand, world undergoing new evolution third generation models: large-scale foundation models. What current state research PT-PHM? trends challenges where do we need go management? This paper provides comprehensive review art analysing more than 200 papers, mostly journals. Some elements guide given end document.

Язык: Английский

---

Remaining useful life prediction methods of equipment components based on deep learning for sustainable manufacturing: a literature review

Artificial intelligence for engineering design analysis and manufacturing, Год журнала: 2025, Номер 39

Опубликована: Янв. 1, 2025

Abstract The operational reliability of large mechanical equipment is typically influenced by the functional effectiveness key components. Consequently, prompt repair before their failure necessary to ensure dependability equipment. prognostic and health management (PHM) technology could track system’s state timely detect faults. Therefore, remaining useful life (RUL) prediction as one components PHM rather important. Accurate RUL results be data support for condition-based maintenance plans. Also, it increase safety while reducing loss human financial resources meet requirements sustainable manufacturing in Industry 4.0 era. However, with widespread use deep learning field intelligent manufacturing, there a lack review on based learning. In this paper, different learning-based methods are summarized classified, along pros cons. Then, case study C-MAPSS dataset mainly conducted compared. And finally, difficulties future directions practical scenarios discussed.

Язык: Английский

---

Aero-Engine Borescope Image Defect Detection Algorithm Using Symmetric Feature Extraction and State Space Model

Symmetry, Год журнала: 2025, Номер 17(3), С. 384 - 384

Опубликована: Март 3, 2025

Enhancing the effectiveness of aviation engine borescope inspection is critical for flight safety. Statistics indicate that defects contribute to 20% mechanical-related accidents, while existing defect detection and segmentation models images suffer from a low operational efficiency suboptimal accuracy. To address these challenges, this study proposes Visual State Space with Multi-directional Feature Fusion Mamba (VMmamba) model constructs real-world dataset. First, feature compensation module symmetrical diagonal optimization fusion developed enhance representation capabilities, expand receptive fields, improve extraction model. Second, content-aware upsampling introduced restructure contextual information complex scene understanding. Finally, learning process optimized by integrating Smooth L1 Loss Focal strengthen recognition. The experimental results demonstrate VMmamba achieves 43.4% mAP 36.4% on our dataset, outperforming state-of-the-art 2.3% 1.4%, respectively, maintaining 29.2 FPS inference speed. This framework provides an efficient accurate solution analysis, offering significant practical value maintenance safety-critical decision making.

Язык: Английский

---

Bearing Lifespan Reliability Prediction Method Based on Multiscale Feature Extraction and Dual Attention Mechanism

Applied Sciences, Год журнала: 2025, Номер 15(7), С. 3662 - 3662

Опубликована: Март 27, 2025

Accurate prediction of the remaining useful life (RUL) rolling bearings was crucial for ensuring safe operation machinery and reducing maintenance losses. However, due to high nonlinearity complexity mechanical systems, traditional methods failed meet requirements medium- long-term tasks. To address this issue, paper proposed a recurrent neural network with dual attention model. By employing path weight selection methods, Discrete Fourier transform, mechanisms, accuracy generalization ability in complex time series analysis were significantly improved. Evaluation results based on mean absolute error (MAE) root square (RMSE) indicated that mechanism effectively focused key features, optimized feature extraction, improved performance. An end-to-end RUL model established MS-DAN network, effectiveness method validated using IEEE PHM 2012 Data Challenge dataset, providing more accurate decision support equipment engineers.

Язык: Английский

---

Novel formulations and metaheuristic algorithms for predictive maintenance of aircraft engines with remaining useful life prediction

Reliability Engineering & System Safety, Год журнала: 2025, Номер unknown, С. 111064 - 111064

Опубликована: Апрель 1, 2025

Язык: Английский

---

Intelligent Diagnostic System for Mechanical Fault Detection Using Deep Learning

Journal of Machine and Computing, Год журнала: 2025, Номер unknown, С. 831 - 846

Опубликована: Апрель 5, 2025

Advanced diagnostic tools are essential for aerospace transportation systems and automotive industries industrial manufacturing facilities since operational efficiency requirements safety needs demand failure prediction tools. Systems that use traditional methods depend on centralized architectures show limitations regarding scalability while being unable to overcome subsystem events effectively. The research presents Gossip Neural Network (GNN) as a decentralized deep learning (DL) system which determines Remaining Useful Life (RUL) duration in distributed mechanical engine systems. GNN combines Convolutional Networks (CNNs) Long short- term Memory (LSTM) network layers identify short-term sensor anomalies addition capturing long-term degeneration patterns data. A gossip-based protocol allows the facilitate subsystems train shared model together through peer-to-peer collaborations without needing central control. assessment of proposed framework using CMAPSS data proves its exceptional capability RUL alongside reliable accuracy low error rates. demonstrated excellence different datasets R² results between 92.43% 94.57% RMSE within 12.77 12.87 demonstrates effectiveness handling realistic environments. provides an encouraging solution time-sensitive fault detection facilitates efficient predictive maintenance across large engineering applications.

Язык: Английский

---

Dlcformer: A Dual-Channel Lightweight Convolutional Transformer Model for Remaining Useful Life Prediction

Опубликована: Янв. 1, 2025

Язык: Английский

---