Abstract
In the paper the evolutionary strategy is used for learning of neuro-fuzzy structures of a Mamdani type applied to modelling of nonlinear systems. In the process of evolution we determine parameters of fuzzy membership functions, specific t-norm in a fuzzy inference, specific t-norm for aggregation of antecedents in each rule, and specific t-conorm describing an aggregation operator. The method is tested using well known approximation benchmarks. © 2011 Springer-Verlag.
Cite
CITATION STYLE
Cpałka, K., Rebrova, O., Nowicki, R., & Rutkowski, L. (2011). On designing of flexible neuro-fuzzy systems for nonlinear modelling. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6743 LNAI, pp. 147–154). https://doi.org/10.1007/978-3-642-21881-1_24
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
