文献类型：期刊文献

中文题名：智能电网需求侧个性化推荐系统

英文题名：Personalized recommendation system in the demand side of smart grid

作者：王喜宾 文俊浩 廖臣 赵瑞锋

第一作者：王喜宾

机构：[1]贵州理工学院大数据学院,贵阳550003;[2]重庆大学大数据与软件学院,重庆401331;[3]贵州电网公司信息中心,贵阳550005;[4]广东电网公司电力调度控制中心,广州510600

第一机构：贵州理工学院大数据学院

年份：2022

卷号：45

期号：1

起止页码：38-49

中文期刊名：重庆大学学报

外文期刊名：Journal of Chongqing University

收录：CSTPCD;;Scopus;北大核心:【北大核心2020】；CSCD:【CSCD2021_2022】；

基金：国家自然科学基金资助项目(61672117);贵州省科技厅人才项目([2017]5789-21);贵州理工学院高层次人才项目(XJGC20190929)。

语种：中文

中文关键词：个性化推荐;需求侧管理;智能电网;推荐系统;需求侧响应;大数据

外文关键词：personalized recommendation;demand side management;smart grid;recommendation system;demand response;big data

摘要：基于智能电网的双向通信基础设施与先进量测设备,个性化推荐技术从收集的需求侧大数据中获取知识,为优化电网运营提供有力支持,并向终端用户推荐面向能源的产品/服务/建议。研究首先探讨了个性化推荐技术的原理以及在需求侧中引入个性化推荐技术的前景;其次,介绍了实现智能电网需求侧推荐系统的关键技术,并对现有研究工作以及未来构建的需求侧个性化推荐系统进行分析;最后,讨论了实现需求侧个性化推荐系统可能遇到的挑战。
Driven by the climate change and energy shortage, smart grid has obtained rapid growth worldwide as a solution for the sustainable development of human society. How to extract information from the demand side big data and optimize the grid operation based on the two-way communication infrastructure and advanced metering infrastructure of the smart grid has become an important research topic. As a technology of data analysis and information filtering, personalized recommendation technology is expected to support the information retrieval from the grid data, and recommend energy-oriented products/services/suggestions to the end user. This paper firstly introduces the basic principles of personalized recommendation technology as well as the prospect of introducing this technology into the demand side. Then, some key technologies of implementing the personalized recommendation systems in the smart grid are presented. Furthermore, this paper reviews the existing research in this field and discusses some potential and promising demand side recommendation systems in future. Finally, some challenges of practically deploying the personalized recommendation systems in the smart grid are examined.