Precompensation, Approximation and an Inverse Nyquist Array Design Technique Using Plant Step Data Only

摘要

Given a multivariable plant with uncertain or unknown dynamics, the use of step response data in control design is described based upon a simple diagonal model of a precompensated plant. The technique is identical in structure to the inverse Nyquist array design technique with Gershgorin circles replaced by confidence circles deduced from simple graphical operations on plant step data. At no stage is a plant model required. The choice of constant precompensator and simplified model by iterative means is described as a sequence of dual optimization problems.