Wind farm stability margin exhibits time-varying characteristics due to the influence of time-dependent factors such as output power fluctuations and short-circuit ratio (SCR) variations

increasing oscillation risks caused by insufficient stability margins. Therefore

an online adaptive control method for the wind farm stability margin based on a multi-machine autonomous collaboration architecture is proposed

including turbine-side autonomous control and farm-side correction control

in which the former is used for fast response and the latter for improving accuracy. First

a frequency-domain wind farm model that is conducive to control design is established

considering multi-machine interactions. Based on this model and time-frequency domain modal analysis method

the weak mode-oriented damping control loops are introduced

whose parameters are used as margin control variables. Then

stability margin evaluation models for both turbine and farm are established containing time-varying parameters and margin control variables. With these two models

a stability margin autonomous controller and a correction controller are designed. Furthermore

artificial neural networks (ANN)-based stability margin evaluation accelerating method is proposed

enabling the online adaptive margin control. Finally

typical time-dependent operating conditions are tested in a real-time simulation platform

showing that the proposed method can be adopted to effectively enhance the robustness of the stability margin against time-varying factors and helps prevent oscillations.