Fault Warning Method for Offshore Wind Farm based on SCADA System and Neural Network
In response to the rapid development of offshore wind farms,wind farm fault early warning has received widespread attention,accurate and timely monitoring of the operation of wind turbines in offshore wind farms,and the realization of accurate fault early warning is a hot issue in current research.In this study,a generator winding temperature prediction model is established by combining SCADA system and GA-BP neural network,and coupled with gray correlation analysis to screen the input layer data of the neural network model and determine the sensitivity index of the model.Based on the principle of statistics combined with the wind farm evaluation index and sliding window,the offshore wind farm operation warn-ing threshold is calculated.Finally,according to the warning threshold and the evaluation index of the u-nit to determine the operational status of the wind farm unit,put forward an offshore wind farm fault early warning method.The research results show that the model can effectively realize the offshore wind farm fault early warning and provide theoretical basis and technical support for the offshore wind farm fault ear-ly warning method.
万方数据
维普
