Data-driven Convexification for Frequency Nadir Constraint of Unit Commitment

The increasing penetration of the renewable energy sources brings new challenges to frequency security power systems. In order guarantee system security, constraints are incorporated into unit commitment (UC) models. Due non-convex form nadir constraint which makes constrained UC (FCUC) intractable, this letter proposes a revised support vector machine (SVM) based parameter separating plane method convexify it. Based on data-driven convexification method, we obtain tractable FCUC model is formulated as mixed-integer quadratic programming (MIQP) problem. Case studies indicate that proposed can less conservative solution than existing methods with higher efficiency.