摘要
随着用户侧用能需求多元化的发展,多元负荷的超短期预测对于动态的大型综合能源系统的规划和优化至关重要。为此,该文提出一种基于长期和短期时间序列网络的多元负荷超短期预测模型。首先采用卷积神经网络来提取多元负荷之间的局部依赖关系,然后使用长短期记忆网络捕获负荷序列的长期依赖关系,使用具有循环跳过结构的长短期记忆网络充分学习负荷序列的超长期重复模式,最后采用自回归层和全连接层进行组合预测。使用平均绝对百分比误差和均方根误差作为评价指标,利用美国亚利桑那州立大学坦佩校区综合能源系统数据集进行验证,并与3种负荷预测方法比较。实验结果表明,提出的预测模型均优于其他方法且有较高的预测精度。
With the development of diversified energy demand on the user side,ultra short-term load forecasting is very important for the planning and optimization of dynamic large-scale integrated energy system.Therefore,this paper proposes a multivariate ultra short-term load forecasting model based on long-and short-term time-series network.First,convolutional neural network is used to extract the short-term dependence between multivariate loads.Then,the long-short term memory network is used to capture the long term dependence of load sequence,and the ultra long-term repetitive pattern of load sequence is fully studied by using the long-short term memory network with recurrent-skip.Finally,the autoregressive layer and full connection layer are used for combined prediction.mean absolute percentage error(MAPE)and root mean square error(RMSE)are used as evaluation indexes,the data set of the integrated energy system of Arizona State University Tempe campus is used for verification,and the three load forecasting methods are used for comparison.The experimental results show that the prediction models proposed in this paper are better than other methods and have higher prediction accuracy.
作者
鲁斌
霍泽健
俞敏
LU Bin;HUO Zejian;YU Min(Department of Computer Science,North China Electric Power University,Baoding 071000,Hebei Province,China)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2023年第6期2273-2282,共10页
Proceedings of the CSEE
关键词
综合能源系统
超短期
多元负荷预测
循环跳过
自回归
integrated energy system
ultra short-term
multivariate load forecasting
recurrent-skip
autoregressive
