A Novel Integrated Identification Method of Model Structure and Parameters for Drive System

In order to improve performance of a drive system, it is a requisite to identify accurate model structure and parameters. Traditional system structure identification uses input and output data to determine the system structure, with the knowledge of which, the parameters can be identified with the parameter- fitting methods. Aiming at the drive system, a novel integrated identification to the synthesis of model structure and parameters is proposed in this paper. Since motor inertia is often considered as a prior in engineering practice, the identified value and the true physical value are compared to build the evaluation function, so as to deduce the system structure inversely. Based on the known model structure, the parameters can be identified by the corresponding RLS-based method. Simulation results demonstrate the effectiveness of the proposed technique for both one-mass system and two-mass system.