A Fast Impedance Measurement Method for Wind Turbine Generators Based on Transfer Learning

Existing impedance measurement methods for renewable generators face a trade-off between accuracy and time efficiency. Single-sine signal injection provides the highest accuracy but suffers from the lowest efficiency, while wideband signal injection improves efficiency at the cost of accuracy. To address the above limitation, this paper proposes a fast impedance measurement method for wind turbine generators (WTGs) based on transfer learning. This method involves the accurate impedance measurement at pre-selected key frequency points using single-sine signal injection, followed by the rapid generation of wideband impedance characteristics via transfer learning. By combining the precision of single-sine signal injection with the efficiency of transfer learning, the proposed method overcomes the drawbacks of existing approaches. The validity of the proposed method is demonstrated through simulation cases on direct-drive permanent magnet synchronous generator-based WTGs (DDPMSG-WTGs).