电力数据融合:基本概念、抽象化结构、关键技术和应用场景  

Power Data Fusion:Basic Concepts,Abstract Structures,Key Technologies and Application Scenarios

作　　者：王红霞[1] 王波[1] 陈红坤[1] 刘畅[1] 马富齐 罗鹏[1] 杨艳[2] WANG Hongxia;WANG Bo;CHEN Hongkun;LIU Chang;MA Fuqi;LUO Peng;YANG Yan(School of Electrical Engineering and Automation,Wuhan University,Wuhan 430072,China;Tus-Institute for Renewable Energy,Qinghai University,Xining 810016,China)

机构地区：[1]武汉大学电气与自动化学院,湖北武汉430072 [2]青海大学启迪新能源学院,青海西宁810016

出　　处：《供用电》2020年第4期24-32,共9页Distribution & Utilization

基　　金：国家自然科学基金面上基金项目“考虑多控制模式融合的互动式负荷调度方法研究”(51477121);青海大学中青年科研基金项目(2019-QGY-12)。

摘　　要：为给“电力—信息—社会”深度耦合下的电力数据融合提供参考,对电力数据融合的基本概念、抽象化结构、关键技术及应用场景进行综述性讨论。首先讨论了目前电力数据融合的必要性和存在的挑战;其次介绍了其基本概念和3个层次的抽象化融合结构;接下来对包括贝叶斯网络、D-S证据理论及深度学习等常用的关键技术进行了归纳;并从结构层次的角度对故障诊断、负荷预测等应用场景进行了讨论;最后对未来多模态、跨层次、全范围的网状融合结构进行了综合性描述。To provide references to the power data fusion under the deep coupling "power-information-society",the basic concepts,abstract structures,key technologies and application scenarios of power data fusion are summarized and discussed.Firstly,the necessity and challenges of power data fusion are discussed.Secondly,the basic concepts and three levels of abstracted fusion structure are introduced;Next,the commonly used key technologies are summarized including Bayesian network,D-S evidence theory and deep learning,etc.The application scenarios such as fault diagnosis and load forecasting are discussed from the perspective of structure level.Finally,a comprehensive description of the multi-modal,cross-level,full-scale reticulated fusion structure is given.

关 键 词：深度耦合 数据融合 多模态 状态感知 网状融合结构 

分 类 号：TM71[电气工程—电力系统及自动化]