A New Bearing Fault Diagnosis Method Based on Deep Transfer Network and Supervised Joint Matching

IEEE photonics journal, Год журнала: 2024, Номер 16(3), С. 1 - 17

Опубликована: Апрель 22, 2024

In practical industrial environment, variable working condition can result in shifts data distributions, and the labeled fault various conditions is difficult to collect because rotating machines often works normal status, insufficient brings samples imbalance performance degradation of intelligent diagnosis model. To overcome these problems, by integrating superiority deep learning method feature-based transfer method, this work proposes an innovative cross-domain framework based on convolutional neural network supervised joint matching. Firstly, continue wavelet transform used process original bearing vibration signals extract time-frequency images. Secondly, a built way fine-tuning, trained features from different domains. Thirdly, new domain adaptation approach, matching, developed conduct feature distribution matching instance reweighting with consideration maximum marginal criterion. The model then predict labels target domain's data. verify proposed approaches, study uses two distinct datasets pertaining defects for conducting presence balanced imbalanced experimental analysis indicates that designed methods achieve desirable diagnostic accuracy possess robust generalization ability

Язык: Английский

---

Rapid Computation of Survival Signature for Dynamic Fault Tree based on Sequential Binary Decision Diagram and Multidimensional Array

Reliability Engineering & System Safety, Год журнала: 2024, Номер unknown, С. 110552 - 110552

Опубликована: Окт. 1, 2024

Язык: Английский

---

Challenges for AI in Healthcare Systems

Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 165 - 186

Опубликована: Окт. 30, 2024

Abstract This paper overviews the challenges of using artificial intelligence (AI) methods when building healthcare systems, as discussed at AIsola Conference in 2023. It focuses on topics (i) medical data, (ii) decision support, (iii) software engineering for AI-based health (iv) regulatory affairs well (v) privacy-preserving machine learning and highlights importance involved utilizing AI systems.

Язык: Английский

---

A novel dynamic machine learning-based eXplainable fusion monitoring: Application to industrial and chemical processes

Machine Learning Science and Technology, Год журнала: 2024, Номер 6(1), С. 015005 - 015005

Опубликована: Дек. 17, 2024

Abstract The complexity and fusion dynamism of the modern industrial chemical sectors have been increasing with rapid progress IR 4.0–5.0. transformative characteristics Industry 4.0–5.0 not fully explored in terms fundamental importance explainability. Traditional monitoring techniques for automatic anomaly detection, identifying potential variables, root cause analysis fault information are intelligent enough to tackle intricate problems real-time practices sectors. This study presents a novel dynamic machine learning based explainable approach address issues process systems. methodology aims detect faults, identify their key causes feature analyze path propagation time magnitude one variable another impact. proposed using domain multivariate granger-entropy-aided independent component (DICA)—distributed canonical correlation approach, incorporating dynamics wrapping supported delay-signed directed graph. utilized application processes verified continuous stirred tank reactor Tennessee Eastman as practical benchmarks. framework’s validations efficiency evaluated established such classic computed ICA DICA standard model scenarios. outcomes results showed that newly developed strategy is preferable previous approaches regarding explainability robust detection identification actual high FDRs low FARs.

Язык: Английский

---

A predictive model for centerline temperature in electrical cabinet fires

Renewable and Sustainable Energy Reviews, Год журнала: 2024, Номер 211, С. 115303 - 115303

Опубликована: Дек. 31, 2024

---