TR2019-147

Robust Nonlinear State Estimation for Thermal-Fluid Models Using Reduced-Order Models: The Case of the Boussinesq Equations


    •  Benosman, M., Borggaard, J., "Robust Nonlinear State Estimation for Thermal-Fluid Models Using Reduced-Order Models: The Case of the Boussinesq Equations", IEEE Conference on Decision and Control (CDC), DOI: 10.1109/CDC40024.2019.9029593, December 2019, pp. 2157-2162.
      BibTeX TR2019-147 PDF
      • @inproceedings{Benosman2019dec,
      • author = {Benosman, Mouhacine and Borggaard, Jeff},
      • title = {Robust Nonlinear State Estimation for Thermal-Fluid Models Using Reduced-Order Models: The Case of the Boussinesq Equations},
      • booktitle = {IEEE Conference on Decision and Control (CDC)},
      • year = 2019,
      • pages = {2157--2162},
      • month = dec,
      • doi = {10.1109/CDC40024.2019.9029593},
      • url = {https://www.merl.com/publications/TR2019-147}
      • }
  • MERL Contact:
  • Research Area:

    Control

We present a method for designing robust, proper orthogonal decomposition (POD)-based, low-order observers for a class of spectral infinite-dimensional nonlinear systems. Robustness to bounded model uncertainties is incorporated using the Lyapunov reconstruction approach from robust control theory. Furthermore, the proposed methodology includes a data-driven learning algorithm that auto-tunes the observer gains to optimize the performance of the state estimation. A challenging numerical example using the 2D Boussinesq equations demonstrates the effectiveness of the proposed observer.

 

  • Related News & Events