Post-Disturbance Dynamic Frequency Prediction of High Wind Power-Penetrated Power Systems Considering Operating Regions

As the capacity of grid-connected wind power becomes larger, many power-grid guidelines have been drawn up, which stipulate that wind turbine generators (WTGs) should participate in frequency regulation. The strategies of frequency regulation provided by variable-speed WTGs depend on the operating regions. However, this impact is neglected in the process of post-disturbance dynamic frequency prediction. To address the issue, an improved system frequency response (SFR) model is proposed with consideration of appropriate frequency control strategies corresponding to each operating region. Afterwards, to reduce the error caused by the improved SFR model, an error-correction module based on extreme learning machine (ELM) is established. On this basis, a model-driven and data-driven integrated method is formed. Finally, the effectiveness and advantages of the method for post-disturbance dynamic frequency prediction are verified through simulations.