Hybrid control of a mechanical system with nonlinear constraint

  • Authors:

  • Virginia Ecaterina Oltean;Radu Dobrescu;Dan Popescu;Matei Dobrescu

  • Affiliations:

  • Department of Control and Computers, POLITEHNICA University of Bucharest, Bucharest, Romania;Department of Control and Computers, POLITEHNICA University of Bucharest, Bucharest, Romania;Department of Control and Computers, POLITEHNICA University of Bucharest, Bucharest, Romania;Department of Control and Computers, POLITEHNICA University of Bucharest, Bucharest, Romania

  • Venue:
  • ICOSSSE '09 Proceedings of the 8th WSEAS international conference on System science and simulation in engineering
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Electromagnetic valves, like other mechatronic automotive systems, present significant nonlinearities thus requiring specific control strategies. If the nonlinearities are conveniently approximated, then the system dynamics can be converted into a hybrid model, making possible the implementation of fast and adequate control algorithms. Starting from a generic model of a magnetically actuated mass spring damper, with linear continuous dynamics but strongly nonlinear input constraints, this paper introduces a Hybrid Control System approximating model, based on a partitioning of the evolution interval of the mass position variable. A piecewise-constant control law is deduced from the state space partition, and the controlled system behaviour is comparatively discussed for the hybrid and for the initial nonlinear constrained control, respectively, within MATLAB simulation experiments. The approximating constant values are computed, for simplicity, as mean values of the nonlinear control. Also, the logical controlled model is deduced. Simulation experiments reveal that the hybrid approximated model behaves similarly to the original nonlinear model with continuous control.