One of the main difficulties for the stochastic dynamic characteristics of strip mills research is to establish the stochastic excitation model. Based on the stochastic rolling force acquisition data, which is processed into stationary, normal and zero mean from a aluminum hot strip tandem mill, the stochastic excitation power spectral density (PSD) model and ARMA time series model are established. Characteristics of rolling force ARMA time series models are discussed by means of random process theory. The rolling forces PSD function of facilitating engineering application is obtained by utilizing Levenberg-Marquardt combined with generalized global planning algorithm, and the stochastic excitation model is established. So it provides the basis for the prediction of rolling force, identification and parameter estimation of dynamic rigidity, and also has significance for analyzing and researching stochastic dynamics characteristics to the rolling process and the plate and strip rolling mill.