A Weighted Predict-and-Optimize Framework for Power System Operation Considering Varying Impacts of Uncertainty
- 3 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 41 (3) , 2133-2145
Abstract
Prediction deviations of different uncertainties have varying impacts on downstream decision-making. Improving the prediction accuracy of critical uncertainties with significant impacts on decision-making quality yields better optimization results. Motivated by this observation, this paper proposes a novel weighted predict-and-optimize (WPO) framework for decision-making under multiple uncertainties. Specifically, we incorporate an uncertainty-aware weighting mechanism into the predictive model to capture the relative impact of each uncertainty on specific optimization tasks, and introduce a problem-driven prediction loss (PDPL) to quantify the suboptimality of the weighted predictions relative to perfect predictions in downstream optimization. By optimizing the uncertainty weights to minimize the PDPL, the proposed WPO framework enables adaptive assessment of uncertainty impacts and joint learning of prediction and optimization. Furthermore, to facilitate weight optimization, we develop a surrogate model that establishes a direct mapping between the uncertainty weights and the PDPL, where enhanced graph convolutional networks and multi-task learning are adopted for efficient surrogate model construction and training. Numerical experiments on the modified IEEE 33-bus and 123-bus systems demonstrate that the proposed WPO framework outperforms the traditional predict-then-optimize paradigm, reducing the PDPL by an average of 55% within acceptable computational time.Keywords
All Related Versions
Funding Information
- NSFC/RGC Joint Research Scheme (N_CUHK422/25)
- National Natural Science Foundation of China (52037006)
- National Natural Science Foundation of China (52422706)
- China Postdoctoral Science Foundation (2023TQ0169)
This publication has 30 references indexed in Scilit:
- Improving Sequential Market Coordination via Value-Oriented Renewable Energy ForecastingIEEE Transactions on Energy Markets, Policy and Regulation, 2025
- Explainable spatiotemporal multi-task learning for electric vehicle charging demand predictionApplied Energy, 2025
- Real-time hosting capacity assessment for electric vehicles: A sequential forecast-then-optimize methodApplied Energy, 2024
- Review of several key processes in wind power forecasting: Mathematical formulations, scientific problems, and logical relationsApplied Energy, 2024
- Dynamic Loss Weighting for Multiorgan Segmentation in Medical ImagesIEEE Transactions on Neural Networks and Learning Systems, 2023
- Power systems optimization under uncertainty: A review of methods and applicationsElectric Power Systems Research, 2022
- Cost-oriented load forecastingElectric Power Systems Research, 2022
- EnhanceFace: Adaptive Weighted SoftMax Loss for Deep Face RecognitionIEEE Signal Processing Letters, 2021
- Extensive comparison of physical models for photovoltaic power forecastingApplied Energy, 2021
- Stochastic unit commitment in microgrids: Influence of the load forecasting error and the availability of energy storageRenewable Energy, 2020