Adaptive decision-trees-based regional voltage control

The continuing success of synchrophasors has ushered in new subdomains of power system applications for real-time situational awareness, online decision support, and robust system control. In this paper, an adaptive decision-tree-based systematic method for open-loop regional voltage control is developed. This approach employs voltage security assessment method to generate voltage secure and insecure operating conditions for decision tree learning. Parallel decision trees corresponding to a set of control combinations are trained. To guarantee the robustness of decision trees and its capability of tracking the system condition after topology changes, the resulting trained trees are updated periodically in real time using an online boosting method. The performance of this approach is evaluated using IEEE 118 bus system.