Modern electric power systems are so huge and complicated that it is often impossible to study their electromagnetic transient analysis without any proper model which represents the practical power systems. In order to handle this kind of problem, various tool packages are provided and many studies have been performed by researchers in the field of transient phenomenon analyzing in electrical power system. As the interest in smart grid and smart power management system are getting popular, this kind of studies on proper modeling transient in power system are needed for the design of stable power system and secure management of it.
To understand a practical power system properly, especially in transient state, these models should have a characteristics of describing drastic changes in the system. Electrical transient state is lasted for extremely short time and has a wide range of frequency bandwidth, the good model to represent it should be a FDNE (Frequency Dependent Networks Equivalent). In this paper, an equivalent model as an admittance transfer function is provided, which is a rational function and its order could be decided by a new proposed method of GA(Genetic Algorithm). Generally coefficients of the rational functions are fitted by data from electrical machinery such as a generator, transformers, transfer lines or real tiny scale of power system, whereas in this study, a practical power system model, courtesy of ERIK, New Zealand south island, is used for more accurate modeling and simulation to prove the usefulness of the proposed FDNE.
Four different step frequencies reaching up to 5000Hz of FDNE models are designed for comparison to find the better performance model in representing the practical system. For more specific study on low frequency features, 1Hz scanning step FDNE model is designed. Every FDNE system designed is linked to AC-DC power networks, containing HVDC CIGRE model, with fault applied and removed test condition, different simulations are performed to demonstrate this methodology and its high caliber.

Key Words: FDNE (Frequency Dependent Network Equivalent), evolution process, GA (Genetic Algorithms), ARMA, optimal order, phenotype, electromagnetic transient.