Design of back propagation optimized Nagar-Bardini structure-based interval type-2 fuzzy logic systems for fuzzy identification

In recent years, fuzzy identification based on system identification theory has become a hot academic topic. Interval type-2 fuzzy logic systems (IT2 FLSs) have become a rising technology. This paper designs a type of Nagar-Bardini (NB) structure-based singleton IT2 FLSs for fuzzy identification problems. The antecedents of primary membership functions of IT2 FLSs are chosen as Gaussian type-2 primary membership functions with uncertain standard deviations. Then, the back propagation algorithms are used to tune the parameters of IT2 FLSs according to the chain rule of derivation. Compared with the type-1 fuzzy logic systems, simulation studies show that the proposed IT2 FLSs can obtain better abilities of generalization for fuzzy identification problems.

Automated summary

This paper proposes a type of NB structure-based IT2 FLSs, which utilize Gaussian type-2 primary membership functions and back propagation algorithms for parameter tuning. Simulation studies demonstrate that the IT2 FLSs have better generalization abilities for fuzzy identification problems compared to type-1 fuzzy logic systems.