A new controller for nonlinear dynamic systems

The controller uses measured performance and applied parameters to find an unknown optimum.

A new output-feedback extremum seeking controller was developed. It works for pre-stabilized nonlinear dynamic systems with unknown steady-state performance maps.

The controller uses only the measured performance, the applied parameter, and internal states. It uses integral sliding mode to enforce equality between the applied parameter rate and a prescribed nonzero search velocity.

A super-twisting differentiator and rate normalization recover a gradient-sign surrogate. This happens without demodulation or averaging.

A ramp-corrected moving-equilibrium coordinate cancels the first-order ramp-induced plant term. This yields explicit flow, reversal, recovery, and dynamic-remainder bounds.

The controller uses a deadbanded hybrid relay to establish local practical convergence to an explicit neighborhood of the unknown optimum. It was tested against an 80-state tubular-reactor benchmark.

The realization succeeded in all 200 held-out Monte Carlo tests. It attained the smallest upper-tail objective gap under the stated tuning protocol.

Key facts

  • The controller works without demodulation or averaging.
  • It succeeded in all 200 held-out Monte Carlo tests.
  • The system uses a deadbanded hybrid relay for local practical convergence.

All sections

Edition