Transient Stability Assessment of Power System Based on Enhanced Projection Twin SVM

Transient stability assessment (TSA) of power system based on enhanced projection twin support vector machine (SVM) is proposed. The high dimensional binomial optimization problem based on traditional SVM for TSA was transformed into two low dimensional binomial optimization problems, and the improved optimization method of particle swarm was used to optimize the parameters of projection twin SVM to improve the TSA accuracy. Firstly, an initial sample set composed of system features and projection energy function features is constructed to obtain the optimal feature collection through feature compression. Then, the efficiency of transient stability evaluation is improved by solving two low-dimensional binomial optimizations after dimensionality reduction, and the optimal projection axes of the transient stability state are obtained, such as stable and unstable classes. At the same time, improved particle swarm optimization is used for parameter selection to improve accuracy. Finally, the validity and accuracy are verified by the IEEE-39 node system.