Review on Graph Neural Networks for Process Soft Sensor Development, Fault Diagnosis, and Process Monitoring

Industrial & Engineering Chemistry Research, Год журнала: 2025, Номер unknown

Опубликована: Апрель 18, 2025

Язык: Английский

---

Causality-driven sequence segmentation assisted soft sensing for multiphase industrial processes

Neurocomputing, Год журнала: 2025, Номер unknown, С. 129612 - 129612

Опубликована: Фев. 1, 2025

Язык: Английский

---

A Graph-Based Time-Frequency Two-Stream Network for Multistep Prediction of Key Performance Indicators in Industrial Processes

IEEE Transactions on Cybernetics, Год журнала: 2024, Номер 54(11), С. 6867 - 6880

Опубликована: Сен. 4, 2024

Deep learning-based soft sensor modeling methods have been extensively studied and applied to industrial processes in the last decade. However, existing models mainly focus on current step prediction real time ignore multistep advance. In actual applications, compared prediction, it is more useful for on-site workers predict some key performance indicators Nowadays, task still suffers from two issues: 1) complex coupling relationships between process variables 2) long-term dependency learning. To ravel out these problems, this article, we propose a graph-based time-frequency two-stream network achieve prediction. Specifically, multigraph attention layer proposed model dynamical graph perspective. Then, network, multi-GAT used extract time-domain features frequency-domain dependency, respectively. Furthermore, feature fusion module combine kinds of based minimum redundancy maximum correlation learning paradigm. Finally, extensive experiments real-world datasets show that outperforms state-of-the-art models. particular, SOTA method, method has achieved 12.40%, 22.49%, 21.98% improvement RMSE, MAE, MAPE three-step using waste incineration dataset.

Язык: Английский

---

Prediction of Teaching Quality of Open Online Courses based on Weighted Markov Chain

Опубликована: Июнь 14, 2024

Язык: Английский

---

Scattering matrix similarity metric optimization for improved defect characterisation based on dynamic graph attention networks

NDT & E International, Год журнала: 2024, Номер 148, С. 103220 - 103220

Опубликована: Авг. 28, 2024

Язык: Английский

---