居民用户需求响应行为影响因素辨识与响应特性提取

发布时间：2018-05-20 16:47

本文选题：数据挖掘 + 需求响应　；参考：《电网技术》2017年07期

【摘要】：制定基于分时电价的需求响应激励策略需要充分掌握居民用户的用电特性。针对此问题,该文建立了基于蒙特卡洛方法的需求响应行为影响因素评估模型,基于此模型分析居民用户用电数据和调查问卷,得到了居民用户需求响应行为的主要影响因素。基于主成分分析和逐步线性回归建立了用电设备负荷曲线提取算法,剥离了与需求响应行为有较强相关性用电设备负荷曲线,提取了居民用户对分时电价机制的响应特性。与测量数据的对比表明,负荷曲线提取算法是准确和有效的。基于负荷曲线提取算法分析分时电价机制下负荷曲线的演变特性,可为峰时电价制定、需求响应等提供有效的数据支撑。
[Abstract]:It is necessary to fully grasp the power consumption characteristics of the residents to formulate the demand response incentive strategy based on time-sharing price. In order to solve this problem, this paper establishes a demand response behavior assessment model based on Monte Carlo method. Based on this model, the main influencing factors of the demand response behavior of residents are obtained by analyzing the data of household electricity consumption and questionnaire. Based on principal component analysis (PCA) and stepwise linear regression (PCA), a load curve extraction algorithm for electrical equipment is established, which has a strong correlation with demand response behavior, and extracts the response characteristics of residents to time-sharing pricing mechanism. The comparison with the measured data shows that the load curve extraction algorithm is accurate and effective. Based on the algorithm of load curve extraction, the evolution characteristics of load curve under time-sharing price mechanism are analyzed, which can provide effective data support for peak-hour pricing, demand response and so on.
【作者单位】：大数据工程技术研究中心(上海交通大学电子信息与电气工程学院);
【基金】：国家重点研发计划支持项目(2016YFB0900101) 国家自然科学基金项目(51477100)~~
【分类号】：TM73

【相似文献】