Studies on Artificial Neural Network Based Wind Power Grid-connected Harmonic Resonant Frequency Detection

With the wide utilization of wind resources, the penetration of wind power is ever increasing. Due to the influence of various uncertainties in the wind farm, the wind farm may generate harmonic oscillations randomly in the system. Accurate detection of the resonant frequency of the system is an important prerequisite for effective harmonic suppression. In response to the aforementioned issues, this paper proposes a wind power grid-connected harmonic resonant frequency measurement based on artificial neural network. It takes the instantaneous current values measured in a certain power frequency cycle as the model input, and uses the resonant frequency of the system as the model output. The accuracy and applicability of the proposed measurement method are verified by simulation studies in the MATLAB/Simulink. Results show that the detection method proposed in this paper is highly accurate and has the potential to be used in the wind farm.