.png){kind=icon}
.png){kind=icon}
Return
Oriented Data Generation for Power System Transient Stability Boundary Exploration Based on Support Vector Machine
.png){kind=icon}
OA
AI
DOI：10.3390/electronics14071285.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Transient stability boundary is an important tool for the security and stability analysis of power systems, but typically, obtaining it will result in a high computational burden. Existing methods for obtaining transient stability boundary often involve grid scanning, which generates a large number of dense operating points, followed by transient simulation, which incurs substantial time costs. This paper proposes a method for transient stability boundary exploration based on a Support Vector Machine (SVM), which uses a small number of operating points to obtain a more precise stability boundary while efficiently and accurately acquiring stable operating points near the stability boundary. Firstly, the SVM is employed to classify the initial samples and determine the transient stability boundary under the N-1 fault. Secondly, re-sampling is conducted for operating points near the stability boundary determined by the initial samples. After updating the sample set, SVM classification is performed again. This process is iterated multiple times, and the operating points are continuously generated in the direction oriented toward the actual stability boundary while obtaining a more precise transient stability boundary under the N-1 fault. Finally, the proposed method is validated using actual operational data from a regional power grid of China Southern Power Grid (Guangzhou, China), demonstrating the accuracy and efficiency of the proposed approach.
Keywords:
transient stability boundary
support vector machine
re-sampling
oriented data generation
boundary exploration
Journal
IF:
2.6
Papers:
9.6K
Citations:
4.7W