“…To apply the method described in Section IV-B, the system (2) must be linearized. In general, the various proposals SI consider linear equivalents of system (2), claiming that the linear representation is sufficient to represent the underlying nonlinear phenomenon [18], [19]. However, in the case presented, if one computes the gain K i using the method in Section IV-B and then implements the proportional controller in the nonlinear system (2), the control action may fail in maintaining stability because of the lack of representativeness of the linear equivalent, as shown in the scheme in Fig.…”