基于多元变分模态分解的高比例风电电力系统次同步振荡源定位

Accurately and promptly extracting subsynchronous oscillation (SSO) components from measurements and locating SSO sources are crucial for SSO suppression. Existing transient energy flow (TEF) based SSO location methods suffer from low location accuracy and poor robustness. To cope with the shortcoming of the traditional TEF in SSO source location, this paper proposes a multivariate variational mode decomposition (MVMD) based SSO source location method to locate the SSO source from the measurements. Firstly, the multi-channel measurement matrix of each generator, including voltage and current measurements, is formed. Then, the multi-channel intrinsic mode functions (IMFs) are simultaneously decomposed from the formed multi-channel measurement matrix by using the MVMD approach, enabling the simultaneous decomposition of SSO components from measurements. Furthermore, the IMFs associated with the SSO mode are identified according to the Hilbert transform (HT). Using the identified IMFs, the MVMD-based TEF is calculated and the SSO source is located. Finally, the performance of the proposed method is evaluated using the simulation data of the modified 4-machine 11-bus test system and the field measurements from the Guyuan SSO event in the North China region. The results validate the accuracy and effectiveness of the proposed method in the SSO source location.