Risk-Based Distributionally Robust Optimal Power Flow With Dynamic Line Rating

Cheng Wang, Rui Gao, Feng Qiu, Jianhui Wang, Linwei Xin

Abstract

This paper proposes a risk-based data-driven approach to optimal power flow (DROPF) incorporating dynamic line rating (DLR). To address uncertainties in wind generation and line ratings, a distributionally robust framework is employed using a hybrid ambiguity set based on second-order moments and Wasserstein distance. This approach captures correlations between weather-dependent variables and ensures robustness against data perturbations. The model is reformulated into a tractable conic program, with a simplified version proposed for large-scale systems. Simulation results across various test systems, including a 2,736-bus network, demonstrate the effectiveness and scalability of the proposed method.

Introduction: The Challenge of Uncertainty in Power Grids

Efficiently managing power flow is a cornerstone of modern grid operations. Traditional methods often rely on Static Line Ratings (SLR)—fixed, conservative limits on transmission capacity based on worst-case weather scenarios. While safe, SLR often leaves significant capacity untapped. Dynamic Line Rating (DLR) offers a solution by adjusting limits in real-time based on actual weather conditions.

Table of abbreviations used in the DROPF model.

However, DLR introduces significant uncertainty. Both wind power generation and the cooling effects that determine line capacity are stochastic and highly correlated. For instance, high wind speeds simultaneously increase wind farm output and enhance line cooling. Existing optimization techniques like Stochastic Optimization require precise probability distributions, while Adaptive Robust Optimization can be overly pessimistic. This research introduces a "Goldilocks" approach: Risk-Based Distributionally Robust Optimal Power Flow (DROPF).

Methodology: A Distributionally Robust Framework

The core of the proposed approach is a Distributionally Robust Optimization (DRO) framework. Instead of assuming a single probability distribution, the model optimizes for the worst-case distribution within an "ambiguity set"—a family of distributions consistent with historical data.

Hybrid Ambiguity Set

The researchers developed a hybrid ambiguity set that combines:

  1. Wasserstein Distance: Ensures robustness against small perturbations or outliers in the historical data.
  2. Second-Order Moments: Captures essential statistical properties like mean, variance, and the critical correlations between wind generation and line ratings.

Risk-Based Objective and Tractability

To maintain feasibility in complex scenarios, the model embeds penalties for load shedding, wind curtailment, and line overloads directly into the objective function. The resulting "min-max" problem is reformulated into a Semi-Definite Program (SDP). For larger systems where SDPs become computationally prohibitive, the authors provide a simplified Wasserstein-only (W-DROPF) model that reformulates as a faster Quadratic Program (QP).

Experimental Validation

The models were rigorously tested on three standard benchmarks:

  • 5-bus system: Used for initial validation and sensitivity analysis.
  • IEEE 118-bus system: A medium-scale test for scalability.
  • Polish 2736-bus system: A large-scale network representing real-world complexity.
The 5-bus test system used for detailed illustrative examples.

Key Insights and Results

  • Economic Efficiency: Incorporating DLR consistently reduced dispatch costs compared to SLR by expanding the feasible operating region.
  • Robust Performance: In out-of-sample tests, the hybrid W&M-DROPF model outperformed standard Sample Average Approximation (SAA) and moment-only models, particularly when correlations between wind and line ratings were high.
  • Scalability for Large Grids: The simplified W-DROPF model successfully solved the 2,736-bus Polish system in approximately 12 minutes, demonstrating its practical utility for large-scale grid operators.

Conclusion and Future Directions

The proposed risk-based DROPF model provides a robust, data-driven tool for modern power systems, effectively balancing reliability and economic efficiency. While the current model focuses on single-period static optimization, future work will explore multi-period formulations to account for generator ramping and integrate load demand uncertainty into the ambiguity set.

Summary updated

Abstract

In this paper, we propose a risk-based data-driven distributionally robust approach to investigating the optimal power flow with dynamic line rating. The risk terms, including penalties for load shedding, wind generation curtailment and line overload, are embedded into the objective function. To robustify the solution, we consider a distributional uncertainty set based on the second-order moment, that captures the correlation between wind generation outputs and line ratings, and also the Wasserstein distance, that hedges against data perturbations. We show that the proposed model can be reformulated as a convex conic program. Approximations of the proposed model are suggested, which leads to a significant reduction of the number of the constraints. For practical large-scale test systems, a distributionally robust optimal power flow model with Wasserstein-distance-based distributional uncertainty set and its convex reformulation are also provided. Simulation results on the 5-bus, the IEEE 118-bus and the Polish 2736-bus test systems validate the effectiveness of the proposed models.
BibTeX
@article{wang2018risk,
  author = {Cheng Wang and Rui Gao and Feng Qiu and Jianhui Wang and Linwei Xin},
  title = {{Risk-Based Distributionally Robust Optimal Power Flow With Dynamic Line Rating}},
  journal = {IEEE Transactions on Power Systems},
  year = {2018},
  doi = {10.1109/tpwrs.2018.2844356},
}
Figure