Dynamic security region boundary fitting method based on Transfer Learning

With the maturity of wide area measurement system (WAMS) and the development of big data theory, a large amount of data can be used to assess the transient stability of power system. In this paper, we focus on the boundary fitting of the dynamic security region, which can quantitatively give the safety margin of operating points (OPs). Based on transfer learning (TL) technique, the labeled data which is different but related to the target data, can be used to build the model, and label the target data. Finally, support vector machine (SVM) is used to fit the linear dynamic security region boundary (DSRB), based on the labeled target data. Test results have demonstrated the practicability of the proposed method.