Distributionally Robust Chance-Constrained Optimal Transmission Switching for Renewable Integration

Yuqi Zhou, Hao Zhu, Grani A. Hanasusanto

Abstract

Increasing integration of renewable generation and flexible resources poses significant challenges to ensure robustness guarantees in real-time power system decision-making. This work aims to develop a robust optimal transmission switching (OTS) framework that can relieve grid congestion via judicious line switching. We formulate a two-stage distributionally robust chance-constrained (DRCC) problem that can ensure limited constraint violations for any uncertainty distribution within an ambiguity set. The second-stage recourse variables are first represented as linear functions of uncertainty, yielding an equivalent reformulation of linear constraints only. Furthermore, both moment-based (mean-mean absolute deviation) and distance-based (infinity-Wasserstein distance) ambiguity sets lead to equivalent mixed-integer linear program (MILP) formulations. Numerical experiments on the IEEE 14-bus and 118-bus systems have demonstrated the performance improvements of the proposed DRCC-OTS approaches over two CC-OTS counterparts, in terms of guaranteed constraint violations. In particular, the computational efficiency of the moment-based MILP approach, which is scenario-free with fixed problem dimensions, has been confirmed, making it suitable for real-time grid operations.

Introduction

As power grids integrate increasing levels of intermittent wind and solar power, the unpredictability of these sources creates significant operational challenges. Unexpected spikes in renewable generation can lead to transmission line congestion, often forcing operators to "curtail" or waste clean energy to maintain system stability.

Optimal Transmission Switching (OTS) offers a strategic solution by dynamically reconfiguring the grid's topology. By opening or closing specific circuit breakers, operators can reroute power flows to relieve congestion and maximize renewable integration. However, performing OTS under uncertainty is computationally demanding. Traditional stochastic optimization methods often struggle with real-time requirements, while standard robust optimization can be overly conservative. This research addresses the need for a framework that is both mathematically rigorous regarding safety guarantees and computationally efficient for real-time deployment.

Methodology

The proposed framework utilizes a two-stage Distributionally Robust Chance-Constrained (DRCC) model. In the first stage, the operator determines the optimal line switching actions. In the second stage, following the realization of renewable output, the system adjusts generator outputs (recourse actions) to balance the grid.

Linear Decision Rules (LDR)

A significant contribution of this work is the analytical proof that for a DC power flow model, the recourse actions—specifically how line flows and bus angles respond to uncertainty—are exactly linear functions of that uncertainty. While LDR is often used as an approximation in robust optimization, its exactness in the context of DC-OTS allows for a significant simplification of the mathematical model without sacrificing accuracy.

Ambiguity Sets and MILP Reformulation

Rather than assuming a single probability distribution for renewable generation, the model considers a "family" of possible distributions within an ambiguity set. Two types of sets are explored:

  1. Moment-based (Mean-MAD): Utilizes the mean and mean absolute deviation. This leads to a "scenario-free" model with fixed dimensions, ensuring high scalability.
  2. Distance-based ($\infty$-Wasserstein): A data-driven approach that defines the set based on the distance from historical data distributions.

Using duality theory, these complex probabilistic constraints are reformulated into Mixed-Integer Linear Programs (MILP), which can be solved efficiently using industrial solvers like Gurobi.

Results and Discussion

The performance of the DRCC-OTS approaches was evaluated using standard IEEE 14-bus and 118-bus systems, with uncertainty modeled using real-world hourly wind power data from ERCOT (2018–2020).

As shown in the experimental results, the DRCC approaches (A3 and A4) demonstrate superior reliability compared to traditional Sample Average Approximation (SAA) methods (A1). Specifically, the average violation rates for the DRCC models consistently remain below the target thresholds (e.g., 5% or 10%), whereas SAA often fails to meet these guarantees in out-of-sample tests.

Furthermore, the moment-based DRCC-MAD approach (A3) exhibits exceptional computational efficiency. For the 14-bus system, it achieves solution times of less than 0.5 seconds, making it highly suitable for real-time grid operations. The proposed methods also effectively reduce renewable curtailment compared to standard benchmarks, facilitating higher penetration of clean energy into the grid.

Conclusion

This work provides a robust and efficient framework for optimal transmission switching under uncertainty. By leveraging the exactness of linear decision rules in DC-OTS and employing distributionally robust chance constraints, the model ensures grid reliability while maintaining the speed necessary for real-time control.

Future research will focus on extending this framework to natively incorporate non-linear AC power flow constraints. Additionally, exploring the integration of machine learning to predict optimal switching candidates could further enhance the scalability of the integer programming stage for extremely large-scale power systems.

Summary updated

Abstract

Increasing integration of renewable generation poses significant challenges to ensure robustness guarantees in real-time energy system decision-making. This work aims to develop a robust optimal transmission switching (OTS) framework that can effectively relieve grid congestion and mitigate renewable curtailment. We formulate a two-stage distributionally robust chance-constrained (DRCC) problem that assures limited constraint violations for any uncertainty distribution within an ambiguity set. Here, the second-stage recourse variables are represented as linear functions of uncertainty, yielding an equivalent reformulation involving linear constraints only. We utilize moment-based (mean-mean absolute deviation) and distance-based (infinity-Wasserstein distance) ambiguity sets that lead to scalable mixed-integer linear program (MILP) formulations. Numerical experiments on the IEEE 14-bus and 118-bus systems have demonstrated the performance improvements of the proposed DRCC-OTS approaches in terms of guaranteed constraint violations and reduced renewable curtailment. In particular, the computational efficiency of the moment-based MILP approach, which is scenario-free with fixed problem dimensions, has been confirmed, making it suitable for real-time grid operations.
BibTeX
@article{zhou2022distributionally,
  author = {Yuqi Zhou and Hao Zhu and Grani A. Hanasusanto},
  title = {{Distributionally Robust Chance-Constrained Optimal Transmission Switching for Renewable Integration}},
  journal = {IEEE Transactions on Sustainable Energy},
  year = {2022},
  doi = {10.1109/tste.2022.3203669},
}
Figure