The on-line identification of moment of inertia of servo system

The on-line identification of the moment of inertia is researched, in order to improve the performance of servo system. The identification of the moment of inertia based on the recursive forgetting factor least squares(RFFLS), gradient descent (GDM)and model reference adaptive (MRA)methods is designed respectively. And the advantages and disadvantages of the three methods are summarized. The factors affecting the identification accuracy of the algorithm and convergence time are studied by simulation experiment. In physical experiment, the estimated values of RFFLS and GDM converge faster, but they have obviously oscillation. The estimated value of MRA converges slowly, but it has little oscillation. The estimated errors in steady state of RFFLS, GMD and MRA are 11%,24%and 10%. In order to reduce the estimated errors, the friction obtained by friction model is applied to the identification of the moment of inertia. The effectiveness of this method is demonstrated by experimental results.