Traditional approaches for regulating and maintaining system frequency in power transmission systems leverage primary frequency response, automatic generation control (AGC), and regulation services provided by synchronous generators. In the future, on the other hand, distributed energy resources (DERs) at both utility level and in commercial/residential settings are envisioned to complement traditional generation-side capabilities at multiple time scales to aid frequency regulation and maintaining a reliable system operation. Aligned with this emerging vision, this talk considers a distribution system featuring DERs, and presents a system-theoretic optimization strategy for DERs that enables a distribution feeder to emulate a virtual power plant effectively providing services to the main grid at multiple temporal scales. An online distributed algorithm for DERs is designed to enable the active and reactive power at the feeder head to track given setpoints (e.g, dispatch, ramp, or AGC signals), while concurrently ensuring that electrical quantities are within given limits throughout the feeder. The design of the online algorithm leverages primal-dual gradient methods applied to pertinent minimax problems, and its stability is analyzed under a time-varying optimization formalism. The talk will also demonstrates how individual DERs can provide primary frequency response; particularly, power-frequency droop slopes for individual DERs can be designed so that the distribution feeder presents a guaranteed frequency-regulation characteristic at the feeder head.
