智能电网环境下基于居民电力消费行为特征的需求响应策略研究

Demand response (DR) strategies mainly include dynamic pricing and incentive mechanisms. These DR strategies can promote the change of some normal electricity consumption behaviors of end users, improve the efficiency of energy use, and promote power supply and demand balance as well as energy saving and emission reduction. A large amount of residential electricity consumption data are collected in real-time by the smart meters and other data acquisition terminals deployed in the smart grid. Thus, the electricity consumption behavior characteristics of different residential user groups can be discovered from the data. This knowledge is of great significance for the formulation of more effective differentiated DR strategies. Therefore, the main research contents of this project are as follows. First, data preprocessing, clustering analysis and results evaluation methods of the large scale residential electricity consumption data are studied, so as to identify different groups of residential users that have similar electricity consumption patterns. Then the electricity consumption behavior characteristics of different kinds of residential users are discovered. Second, the interaction mechanisms of residential electricity consumption behaviors and DR strategies are explored. Then the dynamic relationship and the interaction of them are comprehensively evaluated. Third, considering the difference of electricity consumption behavior characteristics of different residential users, the corresponding price-based or incentive-based DR strategies and evaluation methods are designed. This research project has a certain theoretical significance and application value for the promotion of power supply and demand balance, the enhancement of power system reliability, the improvement of energy utilization efficiency, and the realization of energy saving and emission reduction targets in China’s smart grid environment.

利用价格信号或激励机制等需求响应策略，能够促使终端用户改变原有的一些电力消费行为，提高能源利用效率，促进电力供需平衡和节能减排等。智能电网中部署的智能电表和其它采集终端实时采集到大量居民用户的电力消费数据，从中能够挖掘不同类型居民用户群的电力消费行为特征，这对于制定更有效的差异化需求响应策略具有重要意义。因此，本项目主要研究：（1）大规模居民电力消费数据的预处理、聚类分析和结果评价方法，以发现具有相似电力消费行为特征的居民用户群，并挖掘不同类型用户群的电力消费行为特征；（2）居民电力消费行为与需求响应策略之间的相互作用机理，并对二者的动态关系和相互作用进行综合评价；（3）考虑不同类型居民用户群电力消费行为特征差异性的价格或激励需求响应策略及其评价方法。本项目的研究对于我国智能电网环境下促进电力供需平衡、增强电力系统可靠性、提高能源利用效率和实现节能减排目标等具有一定的理论意义与应用价值。

智能电网中部署的智能电表和其它采集终端实时采集到大量居民用户的电力消费数据，从中能够挖掘不同类型居民用户群的电力消费行为特征，这对于制定差异化的需求响应策略具有重要意义。为此，本项目主要研究了居民电力消费行为特征挖掘的数据分析方法、居民电力消费行为与需求响应策略的相互作用机理以及考虑居民电力消费行为特征差异性的需求响应策略。围绕智能电网环境下居民电力消费行为特征挖掘的数据分析方法研究，提出了数据分布视角的模糊聚类均匀效应建模方法，构建了基于模糊聚类模型和形状聚类的电力消费模式识别方法，建立了一种短期负荷概率密度预测的深度学习模型，设计了智能电表数据压缩方法和系统；围绕智能电网环境下居民电力消费行为与需求响应策略的相互作用机理研究，提出了面向需求响应和智慧家居能源管理的电器优化调度模型和方法，建立了电动汽车接入微电网的充放电优化调度模型和方法，设计了需求响应环境下面向工业应用的负荷优化调度建模方法；围绕智能电网环境下考虑居民电力消费行为特征差异性的需求响应策略研究，提出了基于元分析的居民电力需求价格弹性和收入弹性度量方法，构建了居民电价方案设计和用户选择决策的多智能体仿真模型，设计了基于非线性电价需求响应结构模型的分时电价评价方法，建立了智能电网环境下基于云的激励需求响应方法和系统。本项目的研究对于提高电力能源利用效率、增强电力系统可靠性和推动能源服务模式创新具有一定的理论和现实意义。通过本项目研究，共发表SCI/SSCI检索期刊论文36篇，在Elsevier出版社出版学术著作章节1章，授权国家发明专利5项，授权美国发明专利2项，登记软件著作权3项，支持了8名研究生开展研究工作。

内容获取失败，请点击重试