Frequency Stability in Low-Inertia Power Systems

Thesis (Ph.D.)--University of Washington, 2020The increased penetration of renewable energy resources particularly those connected via inverters to the electric grid like wind and solar, has resulted in the displacement of traditional synchronous generators. This has subsequently led to a decline in the available rotational inertia from these synchronous generators that provides immediate frequency response in the event of a disturbance to the grid. The result is a larger increase in the frequency deviation, rate of change of frequency, and a slower settling time, all of which can lead to frequency instability and an eventual collapse of the grid. This network condition has been termed low-inertia power systems. The aim of this dissertation is to design control and optimization algorithms that will enable these inverter-based resources participate effectively and optimally in providing frequency control response in a low-inertia power systems by controlling their inverter interfaces. The first part of this dissertation focuses on optimizing the performance of the popular virtual synchronous machine control structure for inverter-based resources, by developing an algorithm to optimally design the inertia and damping gain coefficient of its frequency control loop. This enables these inverter-based resources to participate efficiently in the inertia response portion of primary frequency control, by producing active power proportional to frequency measurements, in response to a power imbalance or disturbance to the grid, just like a synchronous generator. The second part of this dissertation focuses on designing a novel inverter-based resource control strategy termed inverter power control, which is based on model predictive control, to directly determine the optimal active-power set-point for the inverter-based resources in the event of a power imbalance or disturbance in the system. This proposed control framework alleviates a fundamental drawback of the virtual synchronous machine approach that constrains the inverter-based resources to behave like synchronous machines when responding to a frequency event thereby limiting the potentials of these fast acting and flexible inverter-based resources