An optimal design approach for fuzzy systems based on hybrid genetic algorithms

This paper proposes a hierarchical hybrid genetic algorithm (GA) based on an adaptive fuzzy-neural network with varying nodes. This algorithm extracts important rules from a given large rule base to construct an optimal fuzzy model using the GA, and parameters of the model are estimated using a hybrid of the gradient descent and least square estimate in terms of the characteristics of fuzzy systems. The hybrid GA combines the advantages of GA´s strong search capacity and the fast convergence and accuracy of the conventional optimization. Therefore, the algorithm achieves a trade-off between accuracy, reliability and computing time in global optimization. The simulation and application example given demonstrate its effectiveness