Showing papers by "Mojtaba Ahmadieh Khanesar published in 2014"
03 Jul 2014
TL;DR: A novel type reducer for interval type-2 fuzzy systems is proposed which does not require sorting and it is shown that the computational time required by the proposed methods grow linearly as the number of the rules increases.
Abstract: In this paper a novel type reducer for interval type-2 fuzzy systems is proposed. Type reduction of interval type-2 fuzzy systems requires the solution of two nonlinear constrained optimization problems. Existing exact solutions to these problems (Karnik-Mendel algorithms and their variants) require sorting which is known to be computationally very expensive. In this research, these optimization problems are reformulated and novel improved solutions to these problem are proposed which do not require sorting any more. Simulation results show that the results obtained using the proposed methods are exactly the same as that of enhanced Karnik-Mendel algorithms with at least 9% less computational time when the number of rules of fuzzy system is bigger than ten. For more number of rules, it is even possible that the proposed methods converge 37% faster than enhanced Karnik-Mendel algorithms. In addition, it is shown that the computational time required by the proposed methods grow linearly as the number of the rules increases.
3 citations
TL;DR: The proposed algorithm (N3KCA) is similar to what the human brain does, i.e. to predict the new values of the bounds of normative knowledge based on the previous ones and some knowledge, which it has gained from the previous successive updates.
Abstract: This study presents the normative knowledge source for the belief space of cultural algorithm(CA) based on an adaptive Radial Basis Function Neural Network (RBFNN). The use of the RBFNN makes it possible to use the previous upper and lower bounds of the normative knowledge to update them and to extract a logical relationship between the previous parameters of the normative knowledge and their new values. The proposed algorithm(N3KCA) is similar to what the human brain does, i.e. to predict the new values of the bounds of normative knowledge based on the previous ones and some knowledge, which it has gained from the previous successive updates. Finally, the proposed cultural algorithm is evaluated on 10 unimodal and multimodal benchmark functions. The algorithm is compared with several other optimization algorithms including previous version of cultural algorithm. In order to have a fair comparison, the number of cost function evaluation is kept the same for all optimization algorithms. The obtaine...
1 citations