摘要
长期以来,由于低压台区电能计量装置配置不齐、运行数据不易收集,通常采用云计算模型完成低压台区线损率计算,不仅实时性不高而且准确性较低。目前,随着智能电表的推广应用以及智能终端的升级改造,电网边缘节点的数据处理能力正在逐渐提高。基于上述原因,提出了一种基于智能终端的台区线损率计算方案,充分发挥智能终端的边缘计算能力,并通过编程加以实现。采用AP聚类算法,将台区样本分类,聚合分散数据;随后,基于BP神经网络模型拟合样本数据与电气特征参数之间的关系。实验结果表明,所提方案能够缓解云中心的数据处理压力,提高了计算模型的数据处理效率,为运维人员对低压台区的维护和管理提供参考。
For a long time,due to the uneven configuration of low-voltage meter and difficulty in collecting operation data,cloud computing model used to calculate the low-voltage transformer line loss is usually time-consuming and inaccurate.Not only low real-time but also low accuracy.With the popularization and application of smart meters and the rapid advancement of the construction of electricity consumption information collection systems,the real-time and accuracy of line loss management in low-voltage stations are gradually improving.Therefore,smart stations are very suitable as edge nodes for edge computing.This paper proposes a calculation scheme for the line loss rate of the station area based on the edge computing method.The AP clustering algorithm is used to classify the samples of the transformer and aggregate the scattered data.Then,the relationship between the sample data and electrical characteristic parameters is fitted based on the BP neural network model.The experimental results show that the proposed scheme can alleviate the data processing pressure of the cloud center,improve the data processing efficiency of the computing model,and provide a reference for the maintenance and management of the low-voltage station area.
出处
《科技创新与应用》
2021年第33期6-11,共6页
Technology Innovation and Application
基金
内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司科技项目(编号:510141200039)。
关键词
台区线损率
智能终端
边缘计算
BP神经网络模型
transformer line loss
intelligent terminal
edge computing
BP neural network model