Data-driven power system stability assessment

• Main Author: Kang, Hongyu
• Other Authors: Xu Yan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Power system stability assessment
• Online Access: https://hdl.handle.net/10356/177233
• Institution: Nanyang Technological University

With the increasing energy demand and environmental problems, the stability assessment of power systems has become a topic of great concern. In this paper, I utilized data-driven algorithms to conduct an in-depth study of power system stability problems. In addition, I constructed a stability assessment model based on DT, SVM and ANN using Python. Then I trained and validated the model with a real power system dataset. Through experiments and data results visualization, the accuracy of DT is obtained as 0.9953, SVM as 0.9967 and ANN as 0.9968. Meanwhile, the ROC and AUC curves of ANN and SVM are close to the upper-left intersection, which proves that the model achievement is good. By evaluating and comparing the performance of the models, I found that the proposed models all have good generalization ability and accuracy in forecasting the stability of the power system, with ANN having the best model fit.