Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Two-Stage Stochastic Optimal Power Flow for Microgrids With Uncertain Wildfire Effects

Sifat Chowdhury, Yu Zhang

TL;DR

This work tackles the resilience of distribution grids under wildfire risk by developing a two-stage stochastic optimal power flow for microgrids. It integrates wildfire propagation and smoke-induced PV loss into MG operation, using a second-order cone relaxation of DistFlow and separating dynamic line rating to convexify and accelerate solving. The approach introduces a wildfire-aware scenario generation, a data-driven PV smoke model, and simple resistive heat gain approximations, validated on a modified IEEE 22-bus system with 50 scenarios. The results show improved load-serving capability and reduced computational burden, highlighting the method's practical utility for resilient MG deployment during extreme events.

Abstract

Large-scale power outages caused by extreme weather events are one of the major factors weakening grid resilience. In order to prevent the critical infrastructure from cascading failure, power lines are often proactively de-energized under the threat of a progressing wildfire. In this context, the potential of microgrid (MG) functioning in islanded mode can be exploited to enhance the resiliency of the power grid. However, there are numerous uncertainties originating from these types of events and an accurate modeling of the MG is required to harness its full potential. In this paper, we consider the uncertainty in line outages depending on fire propagation and reduced solar power generation due to the particulate matter in wildfire smoke. We formulate a two-stage stochastic MG optimal power flow problem by utilizing a second-order cone relaxation of the DistFlow model. Leveraging an effective approximation of the resistive heat gain, we separate the complicating constraints of dynamic line rating from the resulting optimization problem. Extensive simulation results corroborate the merits of our proposed framework, which is tested on a modified IEEE 22-bus system.

Two-Stage Stochastic Optimal Power Flow for Microgrids With Uncertain Wildfire Effects

TL;DR

This work tackles the resilience of distribution grids under wildfire risk by developing a two-stage stochastic optimal power flow for microgrids. It integrates wildfire propagation and smoke-induced PV loss into MG operation, using a second-order cone relaxation of DistFlow and separating dynamic line rating to convexify and accelerate solving. The approach introduces a wildfire-aware scenario generation, a data-driven PV smoke model, and simple resistive heat gain approximations, validated on a modified IEEE 22-bus system with 50 scenarios. The results show improved load-serving capability and reduced computational burden, highlighting the method's practical utility for resilient MG deployment during extreme events.

Abstract

Large-scale power outages caused by extreme weather events are one of the major factors weakening grid resilience. In order to prevent the critical infrastructure from cascading failure, power lines are often proactively de-energized under the threat of a progressing wildfire. In this context, the potential of microgrid (MG) functioning in islanded mode can be exploited to enhance the resiliency of the power grid. However, there are numerous uncertainties originating from these types of events and an accurate modeling of the MG is required to harness its full potential. In this paper, we consider the uncertainty in line outages depending on fire propagation and reduced solar power generation due to the particulate matter in wildfire smoke. We formulate a two-stage stochastic MG optimal power flow problem by utilizing a second-order cone relaxation of the DistFlow model. Leveraging an effective approximation of the resistive heat gain, we separate the complicating constraints of dynamic line rating from the resulting optimization problem. Extensive simulation results corroborate the merits of our proposed framework, which is tested on a modified IEEE 22-bus system.
Paper Structure (27 sections, 20 equations, 9 figures, 11 tables)

This paper contains 27 sections, 20 equations, 9 figures, 11 tables.

Table of Contents

  1. Introduction
  2. Motivation
  3. Literature Review
  4. Our Contributions
  5. System Modeling
  6. Scenario Generation
  7. Wildfire Modelling
  8. Dynamic Line Rating (DLR)
  9. Smoke Effect
  10. Problem Formulation
  11. Objective Function
  12. First Stage Constraints
  13. Second Stage Constraints
  14. Power Flow Constraints
  15. Reactive Power Constraints
  16. ...and 12 more sections

Figures (9)

  • Figure 1: Framework of our proposed model: the generated renewable power, load demand and tie line outage status serve as the inputs of the resulting optimization problem from which the DLR constraints are separated.
  • Figure 2: Outage lines and DER in the simulated system.
  • Figure 3: Fuel expenditure profiles for the QS units.
  • Figure 4: Optimal scheduling of ES and MS charging/discharging power.
  • Figure 5: Optimal real power generation by the MTs.
  • ...and 4 more figures

Theorems & Definitions (1)

  • Remark 1