基于云边协同的新能源功率预测系统研究

Research for renewable energy power prediction system based on cloud edge collaboration

为满足当前多时间尺度新能源功率预测的数据分析计算要求,文中提出了基于云边协同的新能源功率预测系统设计方案。构建了基于云边协同预测系统的总体架构,云端对功率预测任务和服务器资源进行总体管控,边缘靠近任务需求侧分担存储、计算任务,并与云端形成层级关系;采用Docker容器技术实现云边系统资源的高效利用,通过资源配置算法优化系统多服务器的延时和功耗平衡;针对大数据文件的云边交互提出通信优化传输机制以节省延时和带宽。最后通过实例验证了所提系统可满足多时间尺度大规模新能源功率预测任务的需求。

In order to meet the requirement of mass data analysis and calculation for current multi-time scale new energy power forecasting,a new energy power prediction system design scheme based on cloud edge collaboration is proposed in this paper.The overall structure of cloud-side collaborative forecasting system is built.The cloud side controls power forecasting tasks and server resources,while the edge besides task-demand side and shares storage and computing tasks.The edge forms a hierarchical relationship with the cloud.Docker container technology is used to make the use of resources of cloud edge system efficiently,and resource allocation algorithm is used to optimize the delay and power balance of system multi-servers.For cloud-side interaction of large data files,communication optimization transmission mechanism is proposed to save transmission delay and bandwidth occupancy.Finally,an example is provided to verify that the proposed system can meet the needs of large-scale of power forecasting tasks for new energy during multi-time scale.