Abstract: This paper proposes the order and parameter identification of electrohydraulic servo system and its noise model by time series CARMA. The maximum likelihood and BIC criterion are used to estimate the order and parameter of the model. The suborder and delay of the model are determined according to the asymptotically Gaussian distribution of estimated parameters. This method is applied to identifying the order and parameter of electrohydraulic position and force coupled system and its noise model. The correctness of identified model is illustrated by simulation step response in comparison with measured step response. This method can be used to model identification of other multi-variable mechanical system.

Key words: 电液伺服系统, 多变量系统, 极大似然法, 时间序列, 系统辨识