Generator parameter identification using evolutionary programming

This paper presents a new approach of evolutionary programming (EP) to the parameter identification problem. The EP is used to identify the generator parameters based on the measurements of generator outputs which are highly contaminated by noise. The EP is a very powerful search method and can be used for parameter identification of complex systems, by contrast to the conventional techniques such as the extended Kalman filter (EKF) method. Comparison between the two different methodologies, EP and EKF, is presented in the paper to show the potential of applications of the EP to parameter identification and system modelling.