【摘要】 随着能源技术与信息技术深度融合，信息感知网络、传输网络逐渐向电网末端、向源网荷全流程拓展，是构建新型电力能源数字化生态的“新常态”。风力、光伏等新能源大规模并网和分布式电源的海量接入，给电力通信网，尤其是末端通信网带来了压力。在“新常态”下，电力通信网的接入方式及拓扑结构存在倒置的可能，特殊站点会出现通信资源严重短缺。因此，开展通信站点资源动态研判是构建新型电力通信网的有力支撑。本文构建了基于GMM聚类的通信站点资源自动化研判分析模型，利用GMM高斯混合模型聚类概率值最大的簇，确定站点资源临界阈值，基于RPA自动化流程分类电网站点动态变化，研判站点资源需求。研究成果有助于针对性指导通信站点资源自动化研判，提高电力通信网运行可靠性和健壮性。更多还原

【Abstract】 With the deep integration of energy technology and information technology, the information perception network and transmission network are gradually expanding to the end of the power grid, to the whole process of source-grid-load, which is the “new normal” to build a new digital power and energy ecology.The large-scale grid-connected of new energy sources such as wind power and photovoltaics and the massive access of distributed power sources have brought pressure to the power communication network, especially in the terminal power grid.Under the “new normal”,the access method and topology structure of the power communication network have risk of inversion, and there will be serious shortage of communication resource at special sites.Therefore, carrying out dynamic research and judgment on communication site resources is a strong support for building a new power communication network.In this paper, an automatic judgment and analysis model of communication site resources based on GMM clustering is constructed.The GMM Gaussian mixture model is used to cluster the cluster with the largest probability value, to determine the critical threshold of site resources, and to classify the dynamic changes of power grid sites based on the RPA automation process.The research results help to guide the automatic research and judgment of communication site resources, and improve the reliability and robustness of the power communication network.更多还原