基于回声状态网络的电力EPON业务感知技术  

Echo-State-Network based service-awareness technology of EPON for smart power grid

作　　者：李钟 

机构地区：[1]北京大学动力中心,北京100871

出　　处：《电力系统保护与控制》2018年第6期69-73,共5页Power System Protection and Control

基　　金：国家自然科学基金项目(51707197)~~

摘　　要：配用电是智能电网建设的重要环节,智能配用电新业务的多样化对其通信网络的支撑能力提出了更高的要求。为了进一步提高配用电以太无源光网络(EPON)系统对业务的支持匹配能力,提出了一种基于回声状态网络的电力EPON业务感知算法,采用回声状态网络模型对电力通信业务进行识别感知,以实现EPON对电力通信业务的高效匹配。在此基础上通过OLT与ONU的交互,优化电力EPON系统的资源调度与分配能力。以配用电通信EPON作为典型业务,对基于回声状态网络的业务感知算法进行验证。仿真验证表明,该机制能有效改善丢包率和时延等性能,提高了EPON系统对配用电通信本地接入业务的主动支持能力。Aimed to improve the ability of EPON （Ethernet Passive Optical Network） to support multiple services brought by smart power grid, an Echo-State-Network based traffic-awareness algorithm in EPON system is proposed for smart power grid. The proposed mechanism is able to be aware of characteristics of multi-services and to match their requirements with better quality, through the echo-state-network algorithm. Thus, resources allocation optimization of EPON system is enabled by the proposed algorithm through cooperation between OLT and ONUs. Simulation evaluation of this proposed scheme is conducted and takes smart power grid services as typical application. Simulation results show that this proposed approach can greatly improve the supporting ability for multiple services of smart power grid, in terms of packet loss rate and time delay performances.

关 键 词：以太无源光网络 业务感知 回声状态网络 配用电 智能电网 

分 类 号：TM73[电气工程—电力系统及自动化] TM76