摘要
常规的智能电网数据调度与分发方法多数采用大数据技术原理,缺少电网数据区域划分,其调度目标节点选择偏差较大,导致电网数据调度错失率较高,数据调度与分发任务完成效果不佳。基于此,引入边缘计算提出了一种新的调度与快速分发方法。首先,建立边缘计算集群分层模型,划分数据区域,分析集群节点状态参量,选择最优调度目标节点。其次,根据调度任务处理时间对任务有效价值的影响,设计智能电网数据流调度优先级,分配并执行数据调度处理任务。最后,设计数据快速分发机制,避免在调度处理任务执行中,出现节点失效和调度数据丢失的问题。根据实验结果可知,新的方法应用后,即使数据处理任务个数的增加,数据调度错失率也未出现大幅度波动,均控制在0.2%以下。
Conventional smart grid data scheduling and distribution methods mostly use the principle of big data technology,lacking in grid data area division,and their scheduling target node selection deviation is large,resulting in a high rate of grid data scheduling errors and poor performance in completing data scheduling and distribution tasks.Based on this,a new scheduling and fast distribution method is proposed by introducing edge computing.First,a hierarchical model of edge computing cluster is established,data regions are divided,cluster node status parameters are analyzed,and optimal scheduling target nodes are selected.Secondly,scheduling the impact of task processing time on the effective value of tasks,designing smart grid data flow scheduling priorities,assigning and executing data scheduling processing tasks.On this basis,a rapid data distribution mechanism is designed to avoid node failures and scheduling data loss during the execution of scheduling tasks.According to the experimental results,after the application of the new method,as the number of data processing tasks increases,there is no significant fluctuation in the data scheduling error rate,which is below 0.2%.
作者
原静
孙骏
YUAN Jing;SUN Jun(State Grid Beijing Electric Power Company,Beijing 100000,China)
出处
《信息与电脑》
2023年第6期226-229,共4页
Information & Computer
关键词
边缘计算
分发
调度
智能电网
快速
数据
edge computing
distribution
dispatch
smart grid
fast
data
