基于VMD-BA-LSTM的短期风向预测研究

Short-term wind direction prediction research based on VMD-BA-LSTM

准确的风电场风向预测对制定偏航控制策略、提高发电量及风电机组稳定运行具有重要意义。针对风向的随机性和不确定性的特点,提出一种变分模态分解(VMD)和蝙蝠算法(BA)优化长短期记忆(LSTM)神经网络的短期风向预测模型。首先,采用变分模态分解将原始序列分解为多个有限带宽的特征模态分量以降低原始数据的复杂度和非平稳性对预测精度的影响,然后将各分量分别建立BA-LSTM模型进行预测,最后将各分量预测结果叠加得到风向值,结合河北某风场的实测数据进行多时间尺度的风向预测。实验结果表明,本文所提方法相比于LSTM和最小二乘支持向量机(LSSVM)预测方法提高了预测精度,对后续研究偏航系统的最优调节提供了支持。

Accurate wind farms wind direction prediction is important for evaluating yaw control strategies,improving electric energy production capacity and steady operation of wind turbines.According to the randomness and uncertainty of wind direction,a short-term wind direction prediction model is proposed based on variational mode decomposition(VMD)and bat algorithm(BA)optimized long short term memory(LSTM).Firstly,the original sequence is decomposed into multiple finite-band eigen-mode modal components by using VMD to reduce the influence of the complexity and non-stationarity of the original data on the prediction accuracy.Then the BA-LSTM model for each component is established to predict separately,and finally the wind direction values are obtained by superimposing the prediction results of each component.The multi-time-scale wind direction prediction is carried out based on the measured data of a wind field in Hebei Province.Experimental results show that the proposed method improves the prediction accuracy compared with the LSTM and least square support vector machine(LSSVM)prediction methods,which provides support for the subsequent research on the optimal adjustment of the yaw system.