高负载数据大批量接入后的网关状态预测模型构建  

Construction of Gateway State Prediction Model after Mass Access of High-load Data

作　　者：敖知琪 陈禹旭 任昊文 康旖 代昊琦 AO Zhi-qi;CHEN Yu-xu;REN Hao-wen;KANG Yi;DAI Hao-qi(China Southern Power Grid Digital Grid Group Co.,Ltd.,Guangzhou 510700 China)

机构地区：[1]南方电网数字电网集团有限公司,广东广州510700

出　　处：《自动化技术与应用》2025年第3期84-87,共4页Techniques of Automation and Applications

基　　金：南方电网数字电网集团有限公司“南网云平台(三期)建设项目”(JY-KF-03-YP-21-005-TQ);广东省科技项目(202000320001)。

摘　　要：高负载数据大批量接入后网关信道的容量受数据传输速率影响,导致网关状态预测不准确。构建高负载数据大批量接入后的网关状态预测模型。利用概率响应函数,计算各个网关节点的平均传输距离,构建高负载数据传输模型。利用接入信号矢量与发射矢量的相对熵,分析网关信道容量。利用时间重复性原理,计算网关状态的预测概率,采用基于单周期、多周期综合的多阶段预测模式,实现网关状态预测模型的构建。实验结果表明,所构建模型在接入时长为43 s时可以迅速提高网关数据的转换率,并将网关信道容量控制在85 bit/s以上。The capacity of the gateway channel is affected by the data transmission rate after the high load data is accessed in large quantities,resulting in inaccurate gateway state prediction.The prediction model of gateway state after high load data mass access is constructed in this paper.The probability response function is used to calculate the average transmission distance of each gateway node and build a high load data transmission model.The relative entropy of the access signal vector and the transmission vector is used to analyze the gateway channel capacity.Based on the principle of time repeatability,the prediction probability of the gateway state is calculated,and a multi-stage prediction model based on single cycle and multi cycle synthesis is adopted to build the gateway state prediction model.The experimental results show that the model in this paper can rapidly improve the conversion rate of gateway data when the access time is 43 s,and control the gateway channel capacity above 85 bit/s.

关 键 词：网关状态 数据接入 预测模型 数据传输 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP393[自动化与计算机技术—控制科学与工程]