Resilient Operation of Power Distribution Systems Using MPC-Based Critical Service Restoration

Abinet Tesfaye Eseye, Bernard Knueven, Xiangyu Zhang, Matthew Reynolds, Wesley Jones

Research output: NRELPresentation


Power distribution systems are more prone to disruptions and cause most power system outages. We propose a service restoration technique to recover the system service (electricity delivery) following an extreme event-triggered substation outage. The proposed technique considers the problem of controlling distributed energy resources (DERs) of a distribution system with the objective of achieving maximum load pick up while satisfying network flow and voltage constraints. The problem is formulated as a model predictive control (MPC), where a linearized optimal power flow (OPF) model is employed to describe the network. The formulation is augmented with a ramping (up) reserve product for the DERs to ensure an upward monotonic load restoration as time evolves. We perform simulations considering the IEEE 13-bus test feeder integrated with wind, solar, microturbine, and energy storage devices. We demonstrate the efficacy of the devised technique in restoring the system loads monotonically, without shedding previously restored loads. We also show the benefit of co-optimization of power and reserve products for DERs on service restoration. In addition, the capability of the technique in regulating nodal voltages and reducing renewable power curtailment is demonstrated.
Original languageAmerican English
Number of pages18
StatePublished - 2021

Publication series

NamePresented at the Thirteenth Annual IEEE Green Technologies (GreenTech) Conference, 7-9 April 2021

NREL Publication Number

  • NREL/PR-2C00-79530


  • DER
  • distribution system
  • extreme event
  • MPC
  • OPF
  • optimization
  • reserve
  • resiliency
  • service restoration


Dive into the research topics of 'Resilient Operation of Power Distribution Systems Using MPC-Based Critical Service Restoration'. Together they form a unique fingerprint.

Cite this