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

摘要

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.