Modeling DFIG-based system frequency response for frequency trajectory sensitivity analysis

Rasoul Azizipanah-Abarghooee, Mostafa Malekpour, Yue Feng, Vladimir Terzija

Research output: Contribution to journalArticle


This paper presents a trajectory frequency sensitivity analysis of a large-scale power system deploying its simple and reduced order system frequency response (SFR) under sudden load disturbances. By implementing this simple SFR model, the calculation time of the real-time simulation is reduced while maintaining the needed accuracy. The transfer functions and their related equations in the SFR model are firstly analyzed. Then using trajectory sensitivity analysis, the effects of different parameters (eg, reheat time constant, inertia constant, damping factor, governor regulation, etc.) on the system frequency and its rate of change are investigated. Additionally, the doubly fed induction generator (DFIG) based SFR model with typical rate of change of frequency (RoCoF) based inertial control is derived for system frequency dynamic studies in power systems. The analysis is presented as the evolution of the sensitivities for the frequency and RoCoF in time with respect to the SFR's initial conditions helping to determine the importance of each parameter in different time frames. Furthermore, the trajectory sensitivities are used for the estimation of frequency response with perturbed parameters. The system frequency response in case of DFIG is also evaluated. The observations are useful for power system operators to quantify the grid frequency control capability under wind power penetration.
Original languageEnglish
Article numbere2774
Number of pages17
JournalInternational Transactions on Electrical Energy Systems
Issue number4
Early online date3 Dec 2018
Publication statusPublished - 1 Apr 2019


  • dynamic response
  • DFIG
  • frequency
  • rate of change of frequency
  • trajectory sensitivity
  • wind energy


Dive into the research topics of 'Modeling DFIG-based system frequency response for frequency trajectory sensitivity analysis'. Together they form a unique fingerprint.

Cite this