3 MODEL BY T-S FUZZY METHOD In order to solve the issue of huge inference rules duringmultidimensional fuzzy inference process, Takagi and Sugeno suggested the T-S fuzzy model in 1985[10]. T-S fuzzy model has the advantages of simple structure, can approximate any continuous function with arbitrary precision, so it has attracted a lot of attention. The main characteristics of the T-S fuzzy model is the conclusion part consists linear functions. The input space was divided into several fuzzy subspace, first in each fuzzy subspace set up a local linear model, and then use the membership function to connect each local model smoothly, thus to get the global fuzzy model of nonlinear function