Design and application of Nagar-Bardini structure-based interval type-2 fuzzy logic systems optimized with the combination of backpropagation algorithms and recursive least square algorithms

This paper designs a kind of Nagar-Bardini structure based on interval type-2 (IT2) fuzzy logic systems for permanent magnetic drive uncertain parameter forecasting. For each fuzzy rule, the consequent, antecedent and input measurement of IT2 membership functions (MFs) are selected as the Gaussian type-2 MFs with uncertain standard deviations. Backpropagation algorithms and recursive least square algorithms are used to optimize the antecedent, input measurement and consequent parameters of fuzzy logic system (FLS) forecasters, respectively. Compared with the corresponding singleton and non-singleton type-1 (T1) FLSs, two Monte Carlo simulation instances on the basis of data of PMD process show the effectiveness and superiority of non-singleton IT2 FLSs according to the convergence analysis.

Automated summary

This paper presents a Nagar-Bardini structure based on IT2 fuzzy logic systems for forecasting uncertain parameters of permanent magnetic drive. By optimizing the parameters of the fuzzy logic system, the study shows that non-singleton IT2 FLS performs better in convergence analysis.