基于自注意力机制改进的TCN-GRU超短期光伏功率预测  

Improved TCN-GRU UItra Short-term PV PowerPrediction Based on Self-attention Mechanism

作　　者：张贺 郑晓亮[1] ZHANG He;ZHENG Xiao-liang(School of Electrical and Information Engineering,Anhui University of Science and Technology,Huainan 232001,Anhui,China)

机构地区：[1]安徽理工大学电气与信息工程学院,安徽淮南232001

出　　处：《兰州文理学院学报(自然科学版)》2025年第2期84-90,共7页Journal of Lanzhou University of Arts and Science(Natural Sciences)

摘　　要：为提高光伏功率预测精度,提出一种基于自注意力机制改进的TCN和GRU超短期光伏发电功率预测模型.首先,对影响光伏功率的气象因素进行分析;其次,在改变TCN的连接方式的基础上,结合自注意力机制特点优化TCN残差结构及GRU的输出,构建组合预测模型.结果表明该模型具有较好的预测精度.In order to improve the accuracy of photovoltaic power prediction,an ultra-short-term photovoltaic power generation power prediction model based on TCN and GRU improved by self-attention mechanism was proposed.Firstly,the meteorological factors affecting photovoltaic power were analyzed.Then,on the basis of changing the connection mode of TCN,the residual structure of TCN and the output of GRU were optimized by combining the characteristics of the self-attention mechanism to construct a combinatorial prediction model.Finally,combined with experiments,it is shown that the model has good prediction accuracy.

关 键 词：光伏功率预测 自注意力机制 门控循环单元 时间卷积网络 

分 类 号：TD713[矿业工程—矿井通风与安全]