A new fuzzy modeling and identification based on fast-cluster and genetic algorithm
Fucai Liu, Lu Pingli, Run Pei · 2004
A new fuzzy identification algorithm is proposed in this paper, which include five blocks: input variables partition block, fast-cluster block, genetic algorithm block, tuning block and termination block. Fast-cluster block is to identify antecedent parameters of T-S model speedily. Tuning block is to fine-tune the parameters of T-S model using the gradient descent approach and termination block checks if the result is satisfactory. The proposed algorithm not only has the advantage of simplicity, but also has high accuracy, strong automation. The simulations indicate that the algorithm is effective in constructing T-S model for complex nonlinear systems.