.png){kind=icon}
.png){kind=icon}
Return
Learning Multiple Convex Voltage Stability Constraints for Unit Commitment
.png){kind=icon}
PRE
AI
DOI：10.1109/TPWRS.2024.3402081.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing penetration of variable renewable energy (VRE) poses significant challenges to power system voltage stability since VRE units are inverter-based generators (IBG) and often weaken the voltage support of power systems. Therefore, power generation scheduling should avoid operation states that may cause voltage instability issues. This paper proposes a data-driven methodology to learn multiple convex voltage stability constraints that can be effectively embedded into unit commitment. First, we utilize multiple convex polyhedrons to represent nonlinear and nonconvex voltage stability boundaries. Then, the polyhedrons are initialized through decision trees and optimized through a global optimization strategy. Finally, we embed the learned voltage stability boundaries as constraints in unit commitment. Case studies on six high VRE penetrated power systems show the accuracy of constraint learning and the effectiveness of voltage stability improvement.
Keywords:
Power system stability
Stability criteria
Voltage
Numerical stability
Decision trees
Reactive power
Indexes
Static voltage stability
high renewable energy penetration
inverter-based generation
unit commitment
constraint learning
Journal
IF:
7.2
Papers:
1.1W
Citations:
5.0W
Organization
View more organizations