Data-driven power system small-signal stability assessment and correction control model based on XGBoost

The influence of small-signal stability on the safety and stability of the power system is becoming more prominent. A mapping model based on steady-state operation information is established using the sample learning method, which provides a new technical path for the rapid assessment and correction of significant power grid oscillation characteristics. This paper establishes a small signal stability assessment and correction control model based on the Extreme Gradient Boosting (XGBoost) algorithm. Firstly, the XGBoost model is obtained by analyzing the mapping relationship between generator power, node power, branch power, and minimum damping ratio. Then, the sensitivity of the generator damping ratio is calculated, and the objective is to minimize the active power adjustment amount of the generator. The stability constraint and power balance are the constraint conditions to establish the optimization correction model, obtain the optimal adjustment amount, correct the minimum damping ratio, and improve the system's stability. Finally, the minimum damping ratio after correction is obtained, and the modified damping ratio is estimated by XGBoost algorithm. The performance of the proposed model is verified in IEEE 3-machine 9-node and 10-machine 39-node systems. (c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 2021 The 2nd International Conference on Power Engineering, ICPE, 2021.