As the Proton Exchange Membrane (PEM) fuel cell has huge potential market in distributed generation and power source of electrical vehicles due to the characteristics of low operating temperature, good starting performance, high energy density and simple structure[1]. It becomes one of the most promising fuel cells. Especially under today’s energy crisis and environmental pollution, the deeper study on PEM fuel cell is even further important.Fuel cell is a complicated non-linear system with time delay and strong coupling, its working process contains three mixing conditions, solid, liquid and gas, which includes liquid flow, mass transfer, heat transfer and electrochemical dynamic process, so it’s not easy to build a precise mathematical model for the fuel cell system. Heat transfer management is one of the key technologies for PEM fuel cell. There are a few articles about building the model for PEM fuel cell temperature system. Tian[2]built a model for the temperature control system of PEM fuel cell by prediction control theory. Shao[3] built a nonlinear temperature model of PEM fuel cell stack with the large range of disturbances based on the energy conversation, but his non-linear model contains too many uncertain parameters. Hu[4] built a physical dynamic temperature model of PEM fuel cell based on the mass-and-energy balance principle, and designed a 2-dimensional incremental fuzzy controller with integrator for controlling the PEM fuel cell temperature.This paper is trying to identify and forecast the temperature of PEM fuel cell by fuzzy technology. In this model, we’ll divide the fuzzy space evenly by established rules, get the number of fuzzy rules and rules application