This paper presents a dynamic model of Polymer Electrolyte Membrane Fuel Cell (PEM FC) as a MISO system using an identification model. The actual PEMFC system is represented in a non linear mathematical model. By identifying the non linear model with Least Square Method, a linear state space model is generated, with and without compensation vector. Another model is derived from linearization in the operating conditions of PEMFC. The Voltage-Current characteristics of each PEMFC models are generated from simulation results, and are compared. It can be seen that the best model is the linear model with compensation vector, since its characteristic is very similar with the typical characteristic of PEMFC. Its Criterion Function of 0.0142 is the smallest among the other models. The smaller the Criterion Function, the model can represent the actual system more accurate. The resulting model can be used for model-based control system.