Unveiling the Segmented Patterns of Energy Storage via Parametric Grid-resilience Value Curve

作者：Yiyang Song, Tiance Zhang, Jianxiao Wang · 发表于：IEEE Transactions on Industry Applications · 年份：2025 · DOI：10.1109/ias62731.2025.11061659 · 被引用次数：2 · 研究领域：Power Systems and Renewable Energy、Energy Load and Power Forecasting、Optimal Power Flow Distribution

The increasing frequency of extreme weather events makes resilience-oriented preventive measures essential for power grids. Specifically, quickly and accurately configuring energy storage before disasters strike becomes a critical but challenging task. To quantify the impact of energy storage on grid resilience, this paper introduces the concept of the resilience value of energy storage (RVES), defined as the savings in both the value of lost load (VoLL) and system operational costs during disasters. By treating storage capacity as a continuous parameter, we develop the grid-resilience value curve (GRVC), an affine function capturing the relationship between low-dimensional parameters and RVES. The GRVC surface can be analytically represented using the multi-parametric linear programming (MPLP) approach, offering intuitive and valuable sensitivity insights to support pre-event storage configuration. To address the combinatorial challenges in MPLP posed by time-coupled constraints in disaster scenarios, we propose two algorithmic enhancements: an umbrella constraint-based MPLP model to remove redundant constraints, and an accelerated algorithm enabling a warm start for efficient GRVC solution. Case studies on IEEE 9 bus and a real-world 121-bus system under ice storm scenarios validate both the practical relevance of GRVC in the resilient configuration of storage and the improved computational efficiency.