view:34849 Last Update: 2020-2-15
Rashtchi, V.; Rahimpour E; Shahrouzi,H
Application of Adaptive Step Length Bacterial Foraging Algorithm For Parameter Identification of Transformer Detailed Model
Detailed R-C-L-M models of power transformers, which are based on lumped parameters, are used extensively not only for transient analysis inside transformer to determine electrical stresses in windings but also for transient study in power system. Models with less number of elements are more practical for system studies due to lack of heavy computations but decreasing the number of model elements worsens model accuracy. In this paper Imperialistic Competitive Algorithm is used to reduce an R-C-L-M model with 43-nodes of a real 400 kV test object to a 21-nodes one. The accuracy of resulted model is verified by compression of transfer function of transformer resulted from imperialist competitive algorithm and Common Analytical Formula.