There are a lot of methods for identification of the Strejc model parameters. The well known graphical method based on the step response can lead to a large error when the step response is disturbed by a random noise. The integral methods of identification, proposed in the paper, allows a reduction of the influence of the disturbances. The special attention was paid to the analysis of the impact of the noise on the accuracy of parameters identification and ways of their elimination. The investigations were made and they are presented in the paper. The proposed method gives accurate models even in the presence of a significant noise.