Abstract: Due to the strong nonlinear dynamic characteristics of hydraulic sliding bearing rotor system, traditional unbalance identification methods have great limitations when applied to unbalance monitoring of the giant arm centrifuge with hydraulic sliding bearing. A real-time unbalance identification method based on radial bearing reaction was proposed for unbalance monitoring of giant arm centrifuge. First, according to the structural characteristics and unbalance components of arm centrifuge, the relationship between radial bearing reaction and test unbalance was obtained by force analysis. Secondly, the influence of variation of bearing stiffness on radial bearing reaction force was analyzed, and the result shows that the influence of non-linear bearing stiffness on bearing ratio of the bearing was relatively small, which indicated that the proposed unbalance identification method based on radial bearing reaction force was reliable theoretically. Finally, the identification method was preliminarily validated by the bench test. Results show that although sliding bearing rotor system has strong non-linear characteristics, the proposed identification method is effective in unbalance identification within the operating range speed.