Power System Transient Stability Prediction Based on Sample Pre-screening of SVM Algorithm and AdaBoost Algorithm

Abstract

As the stability problem of modern power systems is prominent, the stable operation of the power system is more and more popular in power system analysis. To solve the problem of imbalance between the number of unstable and stable samples in transient stability prediction for power system transient prediction using machine learning methods, the paper proposes a transient stability prediction method for power system based on the SVM (Support Vector Regression) algorithm sample pre-screening and AdaBoost (Adaptive Boosting) algorithm. The SVM algorithm is used to partition the sample space of the training sample set, obtain the hyperplane for partitioning, calculate the distance of each stable sample to the hyperplane and rank them. Stable samples with the same number of unstable samples are selected at certain intervals to form a new training sample set, and the AdaBoost algorithm is used to perform transient stability prediction of the power system. It is demonstrated in the IEEE39 system that the proposed method is more accurate than general machine learning algorithms and commonly used methods to deal with sample data imbalance.