Abstract
In this paper, we propose a general method for nonlinear model reduction and identification, inspired by the concept of subspace identification. We propose to use the artificial neural networks to find a nonlinear projection operator that serves to define the reduced state out of the full state or out of an input-output time series. We investigate the viability of the method for both deterministic and stochastic systems.
| Original language | English |
|---|---|
| Pages (from-to) | 1568-1572 |
| Number of pages | 5 |
| Journal | Proceedings of the American Control Conference |
| Volume | 3 |
| State | Published - 1999 |
| Event | Proceedings of the 1999 American Control Conference (99ACC) - San Diego, CA, USA Duration: 2 Jun 1999 → 4 Jun 1999 |