Design of interval type-2 fuzzy logic systems using dung beetle optimizer for fuzzy identification

Faced with the nonlinearity and uncertainty inherent in industrial data, fuzzy identification has emerged as a focal point in system-modeling research. This article proposes a Nagar–Bardini structured interval type-2 fuzzy logic system identification approach driven by dung beetle optimization. The Nagar–Bardini structured interval type-2 fuzzy logic system employs Gaussian interval type-2 membership functions with uncertain standard deviations and performs defuzzification via a weighted combination of upper- and lower-bound firing strengths, balancing computational simplicity with uncertainty handling. The dung beetle optimization imitates the multi-behavior foraging of dung beetles to jointly optimize the rule base and the footprint of uncertainty parameters, providing a tunable exploration–exploitation trade-off while circumventing the local-optimum drawback of gradient-based methods. The approach is validated on Hammerstein plants, nonlinear difference equations, and a high-order system integrated with linguistic rules. The results demonstrate that the proposed interval type-2 fuzzy logic system paradigm delivers superior generalization for nonlinear system identification under uncertainty.