Identification Of Dynamic Parameters For Active Magnetic Bearings

The dynamic parameters of active magnetic bearings, stiffness and damping, are closely related to the feedback controller parameters, which can be changed easily. Identification of dynamic parameters for active magnetic bearings is the foundation for designing the control system and optimizing the control parameters. The dynamic parameters of active magnetic bearings affect the stability of the magnetic bearing system. Identification of dynamic parameters is the precondition of the dynamic response prediction. Therefore, identification of dynamic parameters is the key point of the research on active magnetic bearings. This paper aims to identify the dynamic parameters of active magnetic bearings.Precise rotor model is the foundation of identification of dynamic parameters. At first,this paper built rigid rotor model and flexible rotor model respectively and studied the applicable speed range of each model. Under the test condition, identification of dynamic parameters for active magnetic bearings will be affected by some unknown disturbing force.This paper studied the influence of different disturbing forces on the identification through finite element analysis. In order to solve this problem, improved identification algorithms which can reduce the identification error are proposed and verified. Then, the dynamic parameters identification test was performed. On magnetic bearing testbed, the rotor unbalance response data were collected. Through improved identification algorithms, identification results of the dynamic parameters are obtained and verified. At last, the online identification algorithm of the dynamic parameters is proposed. Using LabVIEW and MATLAB mixed programming technology, online identification program of the dynamic parameters was developed.In this paper, the research on the identification of dynamic parameters for active magnetic bearings obtained some achievements. Identification algorithms were proposed for both the rigid rotor model and the flexible rotor model. The online identification algorithm developed in this paper can greatly speed up the identification speed on the basis of ensuring the accuracy of the identification. This paper provides a foundation for the application of the identification algorithm in engineering.