摘要
功率预测对提高风电电能质量、减少风电并网时对电网的冲击起着重要作用.针对风电功率数据特征,提出一种基于改进变分模态分解(Improved Variational Mode Decomposition,IVMD)的长短期记忆神经网络(Long-Short Term Memory Network,LSTM)的风电功率预测算法,并利用布谷鸟(Cuckoo Search,CS)算法对LSTM进行超参数寻优.首先,通过相关性分析,对风电场10类数据进行特征筛选,确定与功率相关性最大的两类数据作为模型的输入数据.接着,利用IVMD计算最大包络线峰度,确定变分模态分解的最佳分解参数,将原始风速序列进行分解,得到时间尺度各异的本征模态分量(Intrinsic Mode Functions,IMF).最后,针对LSTM神经网络模型的超参数优化困难、难以得到最优解等问题,提出采用CS算法对关键超参数进行寻优,建立了IVMD-CS-LSTM预测模型,得到了风电功率短期预测结果.采用实际风电场数据对算法进行测试,与常用预测算法比较,预测结果有更高的精度.
Power forecasting technology can improve the power quality of wind power and reduce the adverse impact when wind power is connected to the grid.According to the characteristics of wind power data,a Long Short Term Memory network(LSTM)wind power forecasting algorithm based on Improved Variational Mode Decomposition(IVMD)is proposed,and Cuckoo Search(CS)algorithm is used to optimize the super parame-ters of LSTM.First of all,through correlation analysis,10 different data of wind power are screened for char-acteristics,and two types of data with the highest correlation with wind power are determined as the input of the model.Then,IVMD is used to calculate the maximum envelope kurtosis,determine the best decomposition parameters of the IVMD,decompose the original wind speed series,and obtain the Intrinsic Mode Functions(IMF)with different time scales.Finally,in view of the difficulties in optimizing the super parameters of LSTM neural network model and obtaining the optimal solution,the CS algorithm is used to optimize its key super parameters,and the IVMD-CS-LSTM prediction model is established to obtain the short-term prediction results of wind power.The algorithm is tested with the data of Longyuan Electric Power.Compared with the common forecasting algorithm,the prediction results have higher prediction accuracy.
作者
黄峰
喻跃林
扈菲宇
谢鑫
HUANG Feng;YU Yuein;HU Feiyu;XIE Xin(College of Electrical and Information Engineering,Hunan Institute of Engineering,Xiangtan 411104,China)
出处
《湖南工程学院学报（自然科学版）》
2023年第3期1-7,共7页
Journal of Hunan Institute of Engineering(Natural Science Edition)
基金
湖南省自然科学基金项目(2022JJ50116,2022JJ50014)
长株潭国家自主创新示范区建设专项项目(CG-YB20211030,CG-YB20211025).
