Abstract: Aiming at the problems existing in the application of traditional wind measuring tower in wind farm,this paper carries out wind power prediction research based on wind acoustic radar technology and BP neural network.Firstly,we introduce the advantages of wind detection radar in various scenarios of wind power industry;Secondly,the wind power prediction algorithm integrating empirical mode decomposition and neural network is proposed;Finally,taking the sampling data of the wind tower of the acoustic radar in Anhui Nü erling wind farm as an example,the optimal hidden layer number of BP neural network corresponding to the wind speed of 80 meters,85 meters and 95 meters is verified.On this basis,the effectiveness of wind power prediction is verified based on integrated empirical mode decomposition and neural network prediction algorithm under different wind measurement heights of acoustic radar.