Identification For A Class Of Systems With Colored Noises

System identification theory is a methodology describing various system dynamic characters. It is an efficient tool for system studying, using the tool we can depict the system quantitatively. From the sixties last century, as the rapid development of modern control theory, system identification had been going on a good development. Recently, system identification theory has been successfully applied in some engineering fields. But there are still some difficulties to figure out. The thesis study the identification of systems with colored noises, which totally meet the need of the engineering application, expanding some new identification methods.The thesis presents a finite impulse response (FIR) model methods based on model equivalence principle for a class of generalized output error system, one of the systems with colored noises. The basic idea was to approximate the process model and noise model by using FIR models, and then to obtain a special CARMA model which can be identified by the extended least squares, and finally to determine the parameters of the original systems by means of the model equivalence principle. The simulation results can tell the satisfactory of the parameter estimation.For Box-Jenkins systems with correlated noises, this thesis derives a bias compensation least squares identification method by means of the bias compensation principle, and without the stationary and ergodic assumptions of the inputs. The method can be used to general systems with colored noise. The simulation test analyzed that the characters and using field of BCLS method. Based on the bias compensation principle and pre-filtering idea, this paper derived a bias compensation recursive least squares identification algorithm for output error systems with colored noises. The algorithm proposed realized the recursive computation of the bias compensation methods, and can be on-line implemented. The simulation results confirm the theoretical findings.