Research On Ultra Short Term Wind Speed Prediction Algorithm Based On Wind Acoustic Radar Technology

As one of the most mature and valuable power generation methods in the field of renewable energy,wind power has been widely used in the world.With the rapid development of wind power industry and the increasing annual installed capacity,the proportion of large-scale wind power integration is increasing.However,due to the randomness,volatility and intermittence of wind power generation,the volatility and randomness of wind power in large-scale grid connection seriously affect the power flow balance of power system,even affect the security and reliability of large-scale power grid in the process of dynamic operation.In order to effectively solve the problems of "grid connection difficulty" and "wind abandonment" in the process of large-scale grid connection of wind power,ensure the smooth and reliable grid connection of wind power,the wind power prediction technology can be used to formulate a reasonable dispatching plan for the dispatching department.The research based on wind power prediction technology is of great significance to improve the accuracy of wind power prediction and promote the grid connected consumption of renewable energy.Wind power prediction technology can be divided into direct wind power prediction(with power as the prediction object)and indirect wind power prediction(with wind speed as the prediction object).In this article,the ultra short term wind speed prediction of wind farm is studied based on a variety of intelligent algorithms using wind acoustic radar technology as an effective wind measurement means:(1)The wind measurement principle,advantages and disadvantages of wind tower wind measurement method and wind acoustic radar wind measurement method are compared and analyzed,the application scenarios of wind acoustic radar refined wind energy sensing and detection method in wind power industry are summarized;Several typical wind power forecasting methods are introduced and effective forecasting algorithms are classified;The development status of wind power forecasting at home and abroad is analyzed;(2)An ultra short term wind speed prediction algorithm based on integrated empirical mode decomposition and BP neural network is proposed.The prediction algorithm is verified by the data collected from the acoustic radar wind measurement system of Anhui nverling wind farm.The optimal number of hidden layer neurons of BP neural network corresponding to the wind speed at 10 m,30m,70 m,80m,85 m and 95 m is analyzed,On this basis,the effectiveness of the ultra short term wind speed prediction algorithm based on integrated empirical mode decomposition and BP neural network under different wind height is verified;(3)An ultra short term wind speed prediction algorithm based on the integration of empirical mode decomposition and least squares double support vector regression is proposed.The optimal values of kernel function and penalty function of least squares double support vector regression are obtained by adaptive mutation particle swarm optimization algorithm.In order to verify the effectiveness of the proposed algorithm,the wind speed at 10 m and 70 m of the wind speed measurement system of Anhui nverling wind farm is taken as an example to verify the prediction algorithm,the error results of three prediction models for wind speed at 10 m and 70 m verify the effectiveness,high accuracy and adaptability of the integrated empirical mode decomposition and least squares double support vector regression + adaptive mutation particle swarm optimization algorithm in ultra short term wind speed prediction;(4)Aiming at the problem of low prediction accuracy of mapping relationship and regression equation constructed by a single prediction model,a multi model combination ultra short term wind speed prediction algorithm based on integrated empirical mode decomposition + BP neural network and integrated empirical mode decomposition + least squares double support vector regression is proposed by relying on dynamic weight factor.Among them,the new subsequence is generated by the complexity evaluation of the subsequence obtained by integrating the empirical mode decomposition,and the flow chart of multi model combination ultra short term wind speed prediction algorithm based on dynamic weight is given.In order to verify the effectiveness of the proposed algorithm,the wind speed at 10 m and 70 m of the acoustic radar wind measurement system of Anhui nuerling wind farm is taken as an example to verify six wind speed prediction models,the error results of six wind speed prediction models at 10 m and 70 m verify the effectiveness,high accuracy and adaptability of the integrated empirical mode decomposition + fuzzy entropy +combination prediction algorithm + adaptive mutation particle swarm optimization weight algorithm in ultra short term wind speed prediction.The research results of this paper can provide theoretical support for ultra short-term high-precision wind speed prediction of wind farms,and have certain engineering application value.