Loading...

Design of a robust model predictive controller with reduced computational complexity

Razi, M ; Sharif University of Technology

760 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.isatra.2014.09.008
  3. Abstract:
  4. The practicality of robust model predictive control of systems with model uncertainties depends on the time consumed for solving a defined optimization problem. This paper presents a method for the computational complexity reduction in a robust model predictive control. First a scaled state vector is defined such that the objective function contours in the defined optimization problem become vertical or horizontal ellipses or circles, and then the control input is determined at each sampling time as a state feedback that minimizes the infinite horizon objective function by solving some linear matrix inequalities. The simulation results show that the number of iterations to solve the problem at each sampling interval is reduced while the control performance does not alter noticeably
  5. Keywords:
  6. Computational complexity ; Constraints ; Linear matrix inequality ; Model predictive control ; Optimization ; Linear matrix inequalities ; Problem solving ; Robust control ; Robustness (control systems) ; State feedback ; Uncertainty analysis ; Computational complexity reductions ; Control performance ; Model uncertainties ; Number of iterations ; Objective functions ; Optimization problems ; Robust model predictive control
  7. Source: ISA Transactions ; Volume 53, Issue 6 , 1 November , 2014 , Pages 1754-1759 ; ISSN: 00190578
  8. URL: http://www.sciencedirect.com/science/article/pii/S001905781400233X