Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Electrical Engineering

Major Professor

Seddik M. Djouadi

Committee Members

Kevin Tomsovic, Joe Chow, Xiaopeng Zhao, Dan Wilson


Due to high penetration of renewable energy, converter-interfaced sources are increasing in power systems and degrading the grid frequency response. Synthetic inertia emulation and guaranteed primary frequency response is a challenging task. Still, there is high potential for application of highly controllable converter-interfaced devices to help performance. Renewable energy sources and demand side smart devices also need to be equipped with innovative frequency control approaches that contribute to frequency regulation operations.

First, the wind turbine generator is chosen to represent an example of a converter- interfaced source. An augmented system frequency response model is derived, including the system frequency response model and a reduced-order model of the wind turbine generator representing the supportive active power due to supplementary inputs. An output feedback observer-based control is designed to provide guaranteed frequency performance. System performance is analyzed for different short circuit ratio scenarios where a lower bound to guarantee the performance is obtained.

Second, the load side control for frequency regulation with its challenges is introduced. 5G technology and its potential application in smart grids are analyzed. The effect of communication delays and packet losses on inertia emulation are investigated to show the need of using improved communication infrastructure.

Third, a robust delay compensation for primary frequency control using fast demand response is proposed. Possible system structured uncertainties and communication delays are considered to limit frequency variations using the proposed control approach. An uncertain governor dead-band model is introduced to capture frequency response characteristics. Guaranteed inertial response is achieved and compared with a PI-based Smith predictor controller to show the effectiveness of the proposed method.

Fourth, set theoretic methods for safety verification to provide guaranteed frequency response are introduced. The Barrier certificate approach using a linear programming relaxation by Handelman’s representation is proposed with its application to power systems.

Finally, the Handelman’s based barrier certificate approach for adequate frequency performance is studied. The computational algorithm is provided for the proposed method and validated using power system benchmark case studies with a discussion on a safety supervisory control (SSC).

Files over 3MB may be slow to open. For best results, right-click and select "save as..."