A fuzzy neural network with fuzzy impact grades
Neurocomputing
System modelling using fuzzy numbers
IFSA'03 Proceedings of the 10th international fuzzy systems association World Congress conference on Fuzzy sets and systems
| Hi-index | 0.00 |
The commenters analyze the results published in the above paper by Lo and Chen (ibid. vol.17 (1999)) which concern a new method of stability analysis of Takagi-Sugeno fuzzy systems, and amend the comments given by Johansen et al. (ibid. vol.8 (2000)). It is shown that the computational procedure presented by Lo-Chen is not valid for fuzzy systems where the number of rules is greater than three. In reply, Lo-Chen have modified the original claim and patched up the error to generate a sufficient effect provided that an appropriate switching is taken into account; the switching criterion is not overly restrictive and the nature of fuzzy systems permits such practices. They also point out that a computational procedure for multidimensional cases and a routine construction is readily available in the Matlab package