Application of nonlinear system identification in servo system modeling

Nonlinear system identification is one of the main means of establishing dynamics model of complex electromechanical system. The recursive least-squares parameter estimation algorithm is proposed for a class of Hammerstein equations with colored noise error and output error model. The basic idea of the algorithm is a combination of the auxiliary model identification and decomposition technique, the system is decomposed into two subsystems, each subsystem contains a parameter vector. Based on the auxiliary model and the recursive least square theory, replace the unknown identification model in the information vector intermediate variables with the outputs of the auxiliary model, using the estimated noise item can not be residual instead of in the information vector, which can be used to estimate the system recursive identification of all the parameters, the algorithm is very efficient. A simulation example shows the effectiveness of the proposed algorithm.