Experimental results of evolving Takagi—Sugeno fuzzy models for a nonlinear benchmark

The paper offers evolving Takagi-Sugeno (T-S) fuzzy models for a nonlinear benchmark represented by the pendulum-crane system and focused on the dynamics of pendulum angular position behavior. The rule bases and parameters of T-S fuzzy models are continuously evolved by an online identification algorithm in terms of computing the potentials of new data points. Accepting the pendulum angle as model output, two T-S fuzzy models are derived, viz. with a single input and with three inputs. The experimental results are obtained for a pendulum-crane laboratory equipment system and they show the performance of the proposed evolving T-S fuzzy models.