Decision-calibrated prediction sets for robust power system operations

2026-06-01

AI Summary

This work addresses the challenge that conventional uncertainty set calibration methods based on predictive coverage often yield overly conservative sets, compromising the trade-off between economic efficiency and reliability in power systems. The authors propose a decision-calibrated conditional multivariate prediction set that shifts the calibration objective from predictive coverage to the constraint violation rate of downstream operational decisions. To model contextual information and variable dependencies, they introduce a partial input convex neural network to construct a norm-based scoring function. The resulting prediction set is embedded within an affine recourse robust DC optimal power flow framework, enabling scalable implementation. In a 15-minute-ahead reserve scheduling task, the proposed method confines constraint violation deviations within 3 percentage points of the target level—substantially outperforming traditional approaches (>11 percentage points)—while effectively reducing operational costs.

Abstract

Robust optimization offers a tractable approach to balance operating costs and reliability in power systems dominated by weather-dependent renewable uncertainty, but its performance depends critically on the uncertainty set. Standard data-driven approaches often calibrate uncertainty sets to attain predictive coverage, which can produce unnecessarily large sets and costly operating decisions. In contrast, we introduce decision-calibrated prediction sets and embed them as uncertainty sets in robust optimization problems; these are conditional multivariate prediction sets where calibration is defined in terms of the reliability of downstream decisions, rather than in terms of the coverage. First, we learn these conditional prediction sets as sub-level sets of norm-based score functions represented by partially input-convex neural networks, capturing contextual information and multivariate dependence while preserving convexity and tractability in downstream robust formulations. Second, inspired by conformal risk control, we calibrate a score-threshold parameter that sets the volume of the uncertainty set, thereby controlling the expected violations of downstream operational constraints. We apply our approach to 15-minute-ahead reserve scheduling with network-constrained deliverability, which we formulate as a robust DC optimal power flow problem with affine recourse. Numerical experiments show that decision-calibrated sets attain prescribed constraint-satisfaction targets within about three percentage points, whereas standard coverage-based calibration systematically exceeds these targets by more than eleven percentage points, leading to larger sets and higher operating costs.

Problem

• uncertainty set
• robust optimization
• power system operations
• predictive coverage
• decision reliability

Innovation

• decision-calibrated prediction sets
• robust optimization
• partially input-convex neural networks
• conformal risk control
• uncertainty set calibration