Robust and optimal sparse regression for nonlinear PDE models
| Autor: | Roman O. Grigoriev, Daniel R. Gurevich, Patrick A. K. Reinbold |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Partial differential equation
Computer science Applied Mathematics General Physics and Astronomy Statistical and Nonlinear Physics Regression analysis Dynamical Systems (math.DS) Weak formulation 01 natural sciences 010305 fluids & plasmas Nonlinear system Noise Orders of magnitude (time) Temporal resolution 0103 physical sciences FOS: Mathematics Applied mathematics Mathematics - Dynamical Systems 010306 general physics Scaling Mathematical Physics |
| Zdroj: | Chaos: An Interdisciplinary Journal of Nonlinear Science. 29:103113 |
| ISSN: | 1089-7682 1054-1500 |
| DOI: | 10.1063/1.5120861 |
| Popis: | This paper investigates how models of spatiotemporal dynamics in the form of nonlinear partial differential equations can be identified directly from noisy data using a combination of sparse regression and weak formulation. Using the 4th-order Kuramoto-Sivashinsky equation for illustration, we show how this approach can be optimized in the limits of low and high noise, achieving accuracy that is orders of magnitude better than what existing techniques allow. In particular, we derive the scaling relation between the accuracy of the model, the parameters of the weak formulation, and the properties of the data, such as its spatial and temporal resolution and the level of noise. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|