Cooling fan system has been widely used in industrial equipment such as computation host, product line junction box, manufacturing facilities, CNC machining systems and many other power cooling systems. Those systems usually involve high computation power and unavoidably release high dense heat. In order to guarantee long-term performance and healthy status of the kernel operation systems, surrounding temperature is going be controlled through the cooling fan systems. Once the cooling fan system is malfunctioned or the cooling efficiency is degraded, it could cause damage to the core operation system or lower its performance. Therefore, to prevent this condition, status of the cooling fan systems needs be monitored. In this paper, dynamics modeling and parameter identification of a cooling fan system is presented. Once the parameters can be precisely estimated, it can be extended for on-line health monitoring and diagnosis applications. To this aim, modeling of the cooling fan is firstly derived. In this paper, three discrete approximation models are derived for the parameter estimation. Finally, numerical studies are carried to verify the feasibility of the proposed identification method.