Transient Stability Assessment of a Power System Using Multi-layer SVM Method

With the rapid growth of power systems, more large interconnections and the integration of large renewable energies make the systems more complicated. Therefore, transient stability assessment (TSA) has always been considered as one of the top challenges to ensure the security and operation of power systems. The development of Artificial Intelligence (AI) technologies, such as Artificial Neural Network (ANN) and Support Vector Machine (SVM) have been drawn attentions to the power industry recently. Compared with traditional SVM, this paper presents an advanced TSA system using Multi-layer Support Vector Machine (ML-SVM) method. Basically, a Genetic Algorithm (GA) is used in ML-SVM to identify the valued feature subsets with varying numbers of features which makes full use of the input information. Transient stabilities of the system are determined based on the generator relative rotor angles obtained from the time-domain simulation. Data from the time-domain simulation are used as the inputs for ML-SVM training and testing. Then these trained SVMs are integrated to assess the transient stability of the power system. The simulation results show that the proposed method can reduce the possibility of misclassification of the system. Case study of IEEE 9-bus system on PowerWorld Simulator illustrates the effectiveness of the proposed approach.