摘要
当前远程配电定值整定数据呈现出了高速增长的态势,与此同时,海量数据出现了多样化的数据结构形式。针对这一问题,文中研究基于多维异构数据融合的远程配网定值数据处理技术。首先对采集到的数据进行预处理,通过数据清洗、集成、处理等方法来提高原始数据的采集精度。然后采用RSA算法对数据进行加密处理,确保电力数据的安全、快速传输,再通过正交基前向神经网络算法确定最优的网络隐含层数目及神经元个数。将大样本数据分解为小样本数据并由多个处理器进行处理,从而实现对数据的并行化融合。通过实际算例验证了文中所提方法的有效性。
At present,the setting data of remote distribution presents a trend of high-speed growth.At the same time,the massive data appears a variety of data structures.To solve this problem,this paper studies the fixed value data processing technology of remote distribution network based on multi-dimensional heterogeneous data fusion.Firstly,preprocess the collected data,and improve the collection accuracy of the original data through data cleaning,integration,processing and other methods.Then RSA algorithm is used to encrypt the data to ensure the safe and fast transmission of power data.The optimal number of hidden layers and neurons of the network can be determined by orthogonal basis forward neural network algorithm.The large sample data is decomposed into small sample data and processed by multiple processors to achieve parallel data fusion.The effectiveness of the proposed method is verified by a practical example.
作者
汪娇娇
丁健
王汉军
WANG Jiaojiao;DING Jian;WANG Hanjun(Maoming Power Supply Bureau,Guangdong Power Grid Co.,Ltd.,Maoming 525000,China)
出处
《电子设计工程》
2023年第2期116-120,共5页
Electronic Design Engineering
基金
广东电网技改项目(030900GS62180215)。
关键词
多维异构数据融合
数据处理
远程配网
数据传输
multi-dimensional heterogeneous data fusion
data processing
remote distribution network
data transmission
作者简介
汪娇娇(1983-),男,湖北大悟人,硕士,工程师。研究方向:继电保护、电网规划设计、通信、自动化。
