Abstract
The soft sensor models constructed based on historical data have poor generalization due to the characters of strong non-linearity and time-varying dynamics. Moving window and recursively sample updating online modeling methods can not achieve a balance between accuracy and training speed. Aiming at these problems, a novel online learning neural network (LNN) selects high-quality samples with just-in-time learning (JITL) for modeling. And the local samples could be further determined by principal component analysis (PCA). The LNN model shows better performance but poor stability. Weighted multiple sub models, the hybrid model improves accuracy by covering deficiencies. Additionally, the weights could be developed with mean square error (MSE) of each sub model. And the detailed simulation results verify the superiority of adaptive weighted hybrid model.
Original language | English |
---|---|
Title of host publication | Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings |
Publisher | Springer |
Pages | 233-240 |
Number of pages | 8 |
ISBN (Print) | 9783319466712 |
DOIs | |
Publication status | Published - 2016 |
MoE publication type | A4 Conference publication |
Event | International Conference on Neural Information Processing - Kyoto, Japan Duration: 16 Oct 2016 → 21 Oct 2016 Conference number: 23 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 9948 LNCS |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | International Conference on Neural Information Processing |
---|---|
Abbreviated title | ICONIP |
Country/Territory | Japan |
City | Kyoto |
Period | 16/10/2016 → 21/10/2016 |
Keywords
- Hybrid modeling
- Just-in-time learning
- Learning neural network