Abstract: Based on Takagi-Sugeno fuzzy model,a new modeling method is proposed.Firstly,a local linear clustering algorithm is used to determine the number of fuzzy rules and the T-S fuzzy model initial and last parameters,consequently the corresponding 1st order T-S fuzzy neural network is established.Then a hybrid algorithm consisting of a gradient descent algorithm and recurrent least square algorithm is implemented to tune the network paramters,so as to improve accuracy of the fuzzy modeling.Finally,the effectiveness of the proposed method is demonstrated with two simulation examples.