Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/50934
Title: Study of hybrid learning algorithms for realization of self-constructing fuzzy neural networks
Authors: Liu, Fan.
Keywords: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering
Issue Date: 2012
Abstract: Fuzzy neural networks (FNNs) have learning ability and adaptive capability. Usually, the typical approaches of designing an FNN are to build standard neural networks (NNs) which are designed to approximate a fuzzy algorithm or a process of the fuzzy inference system (FIS) through the structure of NNs. The parameters of FNNs can be updated by using learning algorithms of NNs. Therefore, the human-like thinking and reasoning of fuzzy logic systems and the learning and computational ability of NNs can be combined in FNNs. In this thesis, numerous issues pertaining to design of FNN systems have been addressed. The main concerns are to integrate fuzzy logic with NNs to generate self-constructing FNNs. Fuzzy logic system is a rule-based system which comprises of a set of linguistic rules in the form of “IF-THEN”, whereas fuzzy rules are learnt from human beings. In other words, designing a fuzzy system is a subjective approach which is adopted to express the expert’s knowledge. It is difficult for a designer to examine all the input-output data from a complex system and to find a number of appropriate rules for a fuzzy logic system as there is no formal and effective way of knowledge acquisition. Partitioning the input space and determining the appropriate number of fuzzy rules in fuzzy systems are still open issues. Hence, it is highly desired to develop an objective approach to systematizing design procedures for FNN systems from input-output data.
URI: http://hdl.handle.net/10356/50934
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
TeG0701543C.pdf
  Restricted Access
1.86 MBAdobe PDFView/Open

Page view(s) 50

198
checked on Sep 25, 2020

Download(s) 50

5
checked on Sep 25, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.