本研究以深度学习和边缘计算为基础,详细阐述了电力系统负荷预测的方法,并在此基础上分别建立了基于深度学习理论的电力短期负荷预测模型和基于边缘计算电力负荷预测模型,然后基于MATLAB平台分别对两种模型所选取的数据集进行仿真分析,并以N市某充电站为例,分析对比了两种模型对充电站电力负荷预测的准确性。研究结果表明,深度学习模型输出值与真实值相差较小,相关系数为0.998 99,所建立的深度学习模型在数据特征挖掘和分类方面有绝对优势。
Abstract
This study based on the deep learning and edge calculation,expounds on the power system load forecasting methods,and on this basis,respectively,set up power short-term load forecasting model based on the theory of the deep learning and power load forecasting model based on edge of computing,and then the selected two models respectively based on MATLAB platform by simulating the data set.Taking a charging station in N city as an example,the accuracy of the two models for power load prediction of the charging station is analyzed and compared.The research results show that the output value of the deep learning model has a small difference with the real value,and the correlation coefficient is 0.998 99.The established deep learning model has absolute advantages in data feature mining and classification .
关键词
智能电网 /
短期负荷预测 /
深度学习 /
边缘计算
Key words
smart grid /
short term load forecasting /
deep learning /
edge calculation
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 别朝红,林超凡,李更丰,等.能源转型下弹性电力系统的发展与展望[J].中国电机工程学报, 2020,40(9):3-13.
[2] M Liao,A Chakrabortty.Optimization Algorithms for Catching Data Manipulators in Power System Estimation Loops[J].IEEE Transactions on Control Systems Technology,2019,27(3):1203-1218.
[3] 李国庆,刘钊,金国彬,等.基于随机分布式嵌入框架及BP神经网络的超短期电力负荷预测[J].电网技术,2020,44(2): 53-61.
[4] K Aurangzeb,M Alhussein,K Javed,et al.A Pyramid-CNN Based Deep Learning Model for Power Load Forecasting of Similar-Profile Energy Customers Based on Clustering[J].IEEE Access,2021,6(99):1.
[5] F Han,T Pu,M Li,et al.Short-term Forecasting of Individual Residential Load Based on Deep Learning and K-means Clustering[J].中国电机工程学会电力与能源系统学报(英文),2017,7(2):1-9.
[6] 陈卓,孙龙祥.基于深度学习LSTM网络的短期电力负荷预测方法[J].电子技术,2018,47(1):39-41.
基金
*南方电网公司重点科技项目资助—智能电网省地调度主站边缘计算集群技术研发与工程示范(GXKJXM20190619)。
PDF(1311 KB)
