Training stiff neural ordinary differential equations in data-driven wastewater process modelling
• Introduce a novel normalisation pair method for training of stiff neural ODEs. • Propose incremental training strategy for enhance performance. • Provide a foundation for broad application of neural ODEs.
Authors
- Evina Katsou (ORCID: https://orcid.org/0000-0002-2638-7579)
- Xiangjun Huang (ORCID: https://orcid.org/0000-0001-9020-3490)
- Kyriakos Kandris (ORCID: https://orcid.org/0000-0003-4173-955X)
Institutions
- Imperial College London (GB)
- Brunel University of London (GB)
Publication Details
- Journal
- Journal of Environmental Management
- Published
- 2024-12-30
- DOI
- https://doi.org/10.1016/j.jenvman.2024.123870
- Citations
- 2
- Primary Topic
- Neural Networks and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.69
Funders
- European Commission