Bearing dynamic parameters identification of a flexible rotor-bearing system based on transfer matrix method

Bearing dynamic parameters are important factors governing the vibration characteristics of rotating machinery; however, they are difficult to determine due to limited experimentation, and inaccurate in modelling. In this paper, a parameter identification method is presented to identify the bearing dynamic parameters of a flexible rotor-bearing system. In this method, the parameter identification problem is formulated as an inverse problem. The bearing dynamic parameters have been characterized through minimizing the error squared of the rotor-bearing system unbalance response between the experiment results and the computational ones. The intergeneration projection genetic algorithm is used to minimize the error squared. As an efficient method for analysing the dynamic behaviour of a rotor-bearing system, an improved transfer matrix method has been employed to calculate the unbalance response. This approach has been applied to identify the bearing dynamic parameters of a test rig supported by two anisotropic bearings, according to the unbalance response experimental data. Results indicated that this method could identify the bearing dynamic parameters. It is also robust to the noise effects.