Hybrid forecasting of chaotic dynamical systems
We work on prediction methods for chaotic dynamical systems by integrating knowledge-based models (KBM) and reservoir computing (RC) techniques within the framework of physics-informed machine learning. The study focuses primarily on the Kuramoto-Sivashinsky (KS) equation, a model emblematic of chao...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/175572 |
| _version_ | 1824456660351975424 |
|---|---|
| author | Zhu, Yicheng |
| author2 | Juan-Pablo Ortega Lahuerta |
| author_facet | Juan-Pablo Ortega Lahuerta Zhu, Yicheng |
| author_sort | Zhu, Yicheng |
| collection | NTU |
| description | We work on prediction methods for chaotic dynamical systems by integrating knowledge-based models (KBM) and reservoir computing (RC) techniques within the framework of physics-informed machine learning. The study focuses primarily on the Kuramoto-Sivashinsky (KS) equation, a model emblematic of chaotic systems prevalent in various physical sciences. We investigate the effectiveness of traditional numerical prediction methods alongside the hybrid Method of Experts (MoE) approaches. Our results show that while the stand-alone RC model and the hybrid model with static weights provide valuable predictive capacity, the integrated MoE model, which incorporates a hybrid of KBM and RC using dynamic weight adjustments based on real-time performance evaluations, provides more accurate predictions over longer periods and maintains a high level of distributional accuracy. This provides a novel approach to predicting chaotic behaviour with potential applications in climate science, epidemiology and fluid dynamics. |
| first_indexed | 2025-02-19T03:57:38Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/175572 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:57:38Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1755722024-05-06T15:37:40Z Hybrid forecasting of chaotic dynamical systems Zhu, Yicheng Juan-Pablo Ortega Lahuerta School of Physical and Mathematical Sciences Lyudmila Grigoryeva juan-pablo.ortega@ntu.edu.sg, lyudmila.grigoryeva@unisg.ch Mathematical Sciences We work on prediction methods for chaotic dynamical systems by integrating knowledge-based models (KBM) and reservoir computing (RC) techniques within the framework of physics-informed machine learning. The study focuses primarily on the Kuramoto-Sivashinsky (KS) equation, a model emblematic of chaotic systems prevalent in various physical sciences. We investigate the effectiveness of traditional numerical prediction methods alongside the hybrid Method of Experts (MoE) approaches. Our results show that while the stand-alone RC model and the hybrid model with static weights provide valuable predictive capacity, the integrated MoE model, which incorporates a hybrid of KBM and RC using dynamic weight adjustments based on real-time performance evaluations, provides more accurate predictions over longer periods and maintains a high level of distributional accuracy. This provides a novel approach to predicting chaotic behaviour with potential applications in climate science, epidemiology and fluid dynamics. Bachelor's degree 2024-04-30T01:52:30Z 2024-04-30T01:52:30Z 2024 Final Year Project (FYP) Zhu, Y. (2024). Hybrid forecasting of chaotic dynamical systems. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/175572 https://hdl.handle.net/10356/175572 en MH4900 application/pdf Nanyang Technological University |
| spellingShingle | Mathematical Sciences Zhu, Yicheng Hybrid forecasting of chaotic dynamical systems |
| title | Hybrid forecasting of chaotic dynamical systems |
| title_full | Hybrid forecasting of chaotic dynamical systems |
| title_fullStr | Hybrid forecasting of chaotic dynamical systems |
| title_full_unstemmed | Hybrid forecasting of chaotic dynamical systems |
| title_short | Hybrid forecasting of chaotic dynamical systems |
| title_sort | hybrid forecasting of chaotic dynamical systems |
| topic | Mathematical Sciences |
| url | https://hdl.handle.net/10356/175572 |
| work_keys_str_mv | AT zhuyicheng hybridforecastingofchaoticdynamicalsystems |