摘要
针对现阶段光伏功率预测方法存在的误差大、稳定性差的问题,提出了基于注意力机制的LSTMATTENTION融合神经网络用于对光伏系统的功率进行预测,通过LSTM神经网络来提取光伏系统输出功率时间序列的特征信息,再添加注意力机制以提高预测精度,最后通过澳大利亚中部乌鲁鲁(艾尔斯岩)的分布式光伏电站提供的数据进行训练与验证。结果表明:所提出的LSTM-ATTENTION神经网络预测模型比单一LSTM模型的预测精度提高了50.25%。因此,该方法可以为光伏系统的实际应用提供有力支持与帮助。
With the recent intensification of global climate change,the development and use of clean energy have been highly valued by governments around the world.The large-scale photovoltaic power generation system poses a major challenge to the safety and stability of the power grid.In response to the problem of large errors and poor stability in the current photovoltaic power prediction methods,a LSTM-ATTENTION fusion neural network is proposed based on attention mechanism to predict the power of photovoltaic systems.The LSTM neural network is used to extract feature information from the output power time series of photovoltaic systems,and then an attention mechanism is added to improve the prediction accuracy.Finally,data provided by the distributed photovoltaic power station in Uluru(Ayers Rock),central Australia,are used for training and validation.The results show that the proposed LSTM-attention neural network prediction model improves the prediction accuracy by 50.25%compared with that of a single LSTM model.Therefore,this method can provide strong support and help for the practical application of photovoltaic systems.
作者
李东泽
齐咏生
刘利强
LI Dongze;QI Yongsheng;LIU Liqiang(Institute of Electric Power,Inner Mongolia University of Technology,Hohhot 010080,China;Intelligent Energy Technology and Equipment Engineering Research Centre of Inner Mongolia Autonomous Region Higher Education Institutions,Inner Mongolia University of Technology,Hohhot 010051,China;Engineering Research Center of Large-scale Energy Storage Technology,Ministry of Education,Inner Mongolia University of Technology,Hohhot 010051,China)
出处
《内蒙古工业大学学报（自然科学版）》
2023年第4期350-354,384,共6页
Journal of Inner Mongolia University of Technology：Natural Science Edition
基金
国家自然科学基金项目(62241309)
内蒙古科技计划项目(2021GG0164)
内蒙古自治区自然科学基金项目(2022MS06018)。
关键词
光伏系统
功率预测
LSTM
注意力机制
photovoltaic system
power prediction
LSTM
attention mechanism
