The automatic adjustment system of machine level is studied by the model experiment. Three kinds of the methods are proposed for automatic adjustment operation; the feedback control method, the fixed model method, and the parameter identification method. It is shown that the feedback control method is efficient for the rough adjustment and that the fixed model method and the parameter identification method, on the other hand, prove useful for the precise adjustment. By the fixed model method, the rapid convergence in adjustment is obtained only if the system characteristics are not so far from those assumed on the model. However in the parameter identification method, the rate of convergence in adjustment does not depend on the system characteristics.
