Dynamic Load Modeling And System Parameter Identification Based On The Extended Kalman Filter

Load modeling has been the importance of awareness and attention to the electricity researchers. Load modeling is one of the difficult problems in power system analysis. Based on the basic conception of load modeling, this paper gives a comprehensive and broad expatiation about the status quo, the structure, the operation of electric power systemand the methods of load modeling. And this paper not only describes the classic algorithms, but also introduces the relatively new method of artificial intelligence mainly including the Genetic Arithmetic, the Artificial Neural Network and immunity strategy; furthermore, a detailed presentation is given about the improved algorithm of the methods mentioned, and the comparison of the superiority and inadequate between the two methods and the classic algorithms is illustrated.It is found that the differential model is easier to be identified but physical principle not so clear, while contrary for induction motor model.