Abstract: | A new adaptive method for identification of the critical gain is proposed. Instead of using relay feedback, a saturation nonlinearity and an adjusting gain are inserted in the loop, and the gain is tuned by the difference signal between the input and the output of the saturation function. The gain decreases monotonically and converges to a constant, which can be very close to the critical gain. It is shown that the feedback system does not become unstable by this method. This method can be easily extended to the multi‐input multi‐output plants. |