Modeling and tackling unit commitment constraint screening under uncertainty

Day-ahead unit commitment (UC) is a fundamental task for power system operators, where generator statuses and power dispatch are determined based on the forecasted nodal net demands. The errors inherent in renewables and load forecasting require uncertainty-aware UC solution procedures, which are nontrivial and time-consuming to solve. In this work, we design a novel screening approach under the forecasting uncertainty to achieve both computation efficiency and solution reliability requirements. Our approach accommodates such uncertainties in both chance-constrained (CC) and robust forms (RO), and can greatly reduce the UC instance size by screening out non-binding constraints. To further improve the screening efficiency, we build upon multi-parametric programming (MPP) theory to convert the underlying screening optimization problem into a piecewise affine function. A multi-area screening approach is further developed to handle the computational intractability issues for large-scale problems. We verify the proposed method’s performance on various UC setups and uncertainty situations. Experimental results show that our robust screening procedure can guarantee better feasibility, while the CC screening can produce more efficient reduced models. On average, the screening time for a single line flow constraint can be accelerated by 71.2X to 131.3X using our proposed method. • We consider the unit commitment constraint screening problem under forecasting uncertainty, which holds great potential for accelerating UC solution and providing reliable decisions. • This work takes an early step in extending the classic deterministic screening framework to accommodate uncertainty-aware UC models. Both principled modeling and solving techniques are proposed for constraint screening under uncertainties. • This work improves the screening efficiency by developing novel affine policies to replace the screening model and proposing a multi-area screening approach for large-scale systems.