A5081344897- Fault Detection and Control Systems
- Machine Learning and ELM
- Mineral Processing and Grinding
- Hydrological Forecasting Using AI
- Ship Hydrodynamics and Maneuverability
- Millimeter-Wave Propagation and Modeling
- Indoor and Outdoor Localization Technologies
- Advanced MIMO Systems Optimization
- Advanced Control Systems Optimization
- Metabolomics and Mass Spectrometry Studies
- Spectroscopy and Chemometric Analyses
- Water Quality Monitoring and Analysis
- Advanced Data Processing Techniques
- Computational Drug Discovery Methods
- Domain Adaptation and Few-Shot Learning
- Machine Learning in Materials Science
- Ocean Waves and Remote Sensing
- Advanced Chemical Sensor Technologies
Zhejiang University of Technology
2022-2024
KU Leuven
2024
The prediction performance of data-driven soft sensors for chemical processes with distributed outputs tends to degrade when distribution discrepancies exist. To meet this challenge, an offset compensation Gaussian process regression model is proposed the quality inference outputs. first captures common molecular weight characteristics different Subsequently, product between operating conditions adjusted by mechanism. From a statistical perspective, conditional discrepancy modeling inputs...
Although recent transfer learning soft sensors show promising applications in multigrade chemical processes, good prediction performance mainly relies on available target domain data, which is difficult to achieve for a start-up grade. Additionally, only employing single global model inadequate characterize the inner relationship of process variables. A just-in-time adversarial (JATL) sensing method developed enhance performance. The distribution discrepancies variables between two different...
In chemical processes, reliable soft sensors are generally established by enough labeled data. However, in most multimode the collection of sufficient data is difficult due to high cost and complexity. this work, transductive transfer broad learning (TTBL) proposed for quality prediction. By transferring useful information from related domain, unlabeled prediction domain utilized modeling. First, feature extracted enhancement nodes. The similarity current captured <tex...