This Is AuburnElectronic Theses and Dissertations

Evaluation and Estimation of Frequency Excursion in the Bulk Power System with High Penetration of Inverter-Based Resources

Date

2024-07-30

Author

Seo, Sangwon

Type of Degree

PhD Dissertation

Department

Electrical and Computer Engineering

Restriction Status

EMBARGOED

Restriction Type

Auburn University Users

Date Available

07-30-2025

Abstract

As the penetration level of renewable energy resources increases, the power system analysis of modern power systems becomes more complex due to the dominance of inverter-based resources (IBRs). The high penetration of IBRs leads to different dynamic responses compared to the conventional power system dominated by synchronous generators (SGs). Therefore, the conventional frequency response analysis that targets the SG-based power system needs revision to take account of IBRs' dynamic behavior within the power system. Considering the inertia constants and droop coefficients from the generator and governor data, the inertial power flow (INPF) and governor response power flow (GRPF) are provided in commercial power system software to estimate the frequency response in the phasor domain. However, this functionality is only limited to the conventional power system targeting the dynamics of the synchronous generators, not considering the behaviors of the IBR resources and their interactions with the existing power system components. Therefore, this paper thoroughly investigates the frequency response and dynamic behaviors of the different types of resources with the different ancillary services by running the dynamic simulation using the IBR generic models in the bulk power system. In addition, this work proposes the estimation models of frequency excursion using the investigated system factors that affect the frequency responses under the different penetration levels of the IBR, so the frequency excursion of the bulk power system with high penetration of the IBR can be quickly and accurately estimated and predicted throughout the proposed system frequency model and the machine-learning model. This prediction model can give substantial values to the system operators to prepare power system contingency, provide enhanced system reliability, and optimize the system specification.