二次分解组合 LSTM 的短期风电功率预测模型

随着风电在电力系统中的占比逐步提高，风电功率的精确预测对电力系统的安全稳定运行具有重要意义。然而，风电的随机性和间歇性极大地影响其功率的精确预测。为此，提出二次分解组合长短期记忆(LSTM)的短期风电功率预测模型。首先，采用经验模态分解(EMD)技术将原始风电序列分解为若干固有模态分量；再采用样本熵(SE)技术将各分量重组为高、中、低频3个序列，针对高频模态混叠再次采用麻雀搜索算法-变分模态分解(SSA-VMD)二次分解技术；最后，采用SSA算法对LSTM的参数进行寻优并完成风电功率预测。以湖北省某风电场对所提模型进行验证，并与其他模型进行对比。结果表明，所提模型的平均绝对误差(MAE)为5.79 kW,均方根误差(RMSE)为5.64 kW,平均百分比误差(MAPE)为17.38%,具有更好的预测精度。

Short-term wind power prediction model for quadratic decompsition combined LSTM

With the gradual increase of the proportion of wind power in the power system,the accurate prediction of wind power is of great significance to the safe and stable operation of the power system. However,the random and intermittent nature of wind power greatly affects the accurate prediction of its power.Therefore,this paper proposes a quadratic decomposition combination LSTM short-term wind power prediction model.Firstly,the EMD technique is used to decompose the original wind power series into several intrinsic mode components.Then,SE technique is used to recombine the components into three sequences:High,middle and low frequency.SSA-VMD technique is used for high frequency mode aliasing.Finally,SSA algorithm is used to optimize the parameters of LSTM and wind power prediction is completed.A wind farm in Hubei Province is used to verify the proposed model and compare with other models.The results show that the MAE,RMSE and MAPE of the proposed model are 5.79,5.64 kW and 17.38%,respectively.