摘要
针对现有大多数方法难以充分挖掘出电网数据潜在价值的问题,提出了一种云边协同背景下基于随机森林算法结合BP神经网络的电网数据资产综合处理技术。该技术在靠近电网数据源一侧部署边缘计算节点,以构建云边协同环境下的电网数字化资产管理系统。利用随机森林算法设计分类器完成电网数据类型的划分,并将各类型数据输入至BP神经网络中进行学习,通过不断地迭代优化输出相应的综合处理结果。基于Python平台进行的实验分析结果表明,所提技术的分类准确率均超过了90%,能够有效提升电网数据资产的处理效率。
Aiming at the problem that most existing methods are difficult to mine the potential value of power grid data,an integrated processing technology based on random forest algorithm and BP neural network for power grid data assets under the background of cloud edge collaboration was proposed.The edge computing nodes close to the grid data source were deployed to build a grid digital asset management system in the cloud edge collaborative environment.At the same time,a classifier was designed by using the random forest algorithm to complete the classification of power grid data types,each type of data was input into the BP neural network for learning,and the corresponding comprehensive processing results were output through continuous iterative optimization.The experimental analysis of as-proposed technology based on Python platform shows that the classification accuracy is more than 90%,adequately effective to improve the processing efficiency of power grid data assets.
作者
陈浩敏
梁锦照
马赟
李晋伟
CHEN Haomin;LIANG Jinzhao;MA Yun;LI Jinwei(Technical Standard Service Center,Digital Power Grid Research Institute Co.,Ltd.of China Southern Power Grid,Guangzhou 510663,Guangdong,China)
出处
《沈阳工业大学学报》
CAS
北大核心
2024年第1期54-59,共6页
Journal of Shenyang University of Technology
基金
国家重点研发计划项目(2017YFB0902900)
中国南方电网有限责任公司科技项目(080008KK52200007)。
关键词
云边协同
随机森林算法
BP神经网络
电网数据资产
电网数字化
分类器
数据处理
负荷预测
cloud edge collaboration
random forest algorithm
BP neural network
grid data assets
power grid digitization
classifier
data processing
load forecasting