Type-1 Fuzzy Systems: Design Methods and Case Studies

This chapter focuses first on what exactly “design of a type-1 fuzzy system” means, then provides a tabular way for making the choices that are needed in order to fully specify a type-1 fuzzy system, introduces two approaches to design—the partially dependent approach and the totally independent approach, includes some important design methods, and has five extensive case studies. This chapter covers the following topics: a detailed overview of what designing type-1 fuzzy systems means; eight design methods for designing a type-1 fuzzy system, namely, one-pass (data assignment, WM), clustering using fuzzy c-means, least squares, derivative-based (back-propagation), derivative-free (QPSO), and hybrid (adaptive network fuzzy inference system—ANFIS, and structure identification and feature extraction—SIFE—for TSK systems); and five case studies (forecasting of time series, knowledge mining using surveys, rule-based classification of video traffic, fuzzy logic control, and explainable type-1 fuzzy systems), all of which are re-examined in Chap. 10 . There are two appendixes: Appendix 1 introduces novel Constraints Almost Always Satisfied Parameters (CAASPs) for type-1 fuzzy sets and systems and Appendix 2 provides a proof to a performance improvement theorem. Sixteen examples are used to illustrate important concepts.