6 CONCLUSIONS This paper introduced a simple and effective fuzzy identification algorithm based on T-S fuzzy model for the multivariable system, and used it to build the temperature model for a fuel cell system. Compared with the previous fuzzy model identification methods, it separates the structure identification and parameter identification, and simplifies the identification complexity of conclusion parameters significantly. it’s simple and practical and has high identification accuracy. The whole identification process needs less CPU time compared with fuzzy clustering method and feedback error learning method so makes a good foundation for the on-line controller design of PEM fuel cell system.