• CSCD核心库收录期刊
  • 中文核心期刊
  • 中国科技核心期刊

电力建设 ›› 2019, Vol. 40 ›› Issue (6): 132-138.doi: 10.3969/j.issn.1000-7229.2019.06.015

• 智能电网 • 上一篇    

考虑资源相关性和不确定性的负荷需求响应决策研究

宋杰1,陈振宇2,杨阳3,张凯恒3,张卫国1,朱庆1,徐青山3   

  1. 1.南瑞集团有限公司,南京市 211106;2.国网江苏省电力有限公司,南京市 210029;3.东南大学,南京市 211189
  • 出版日期:2019-06-01
  • 作者简介:宋杰(1981),男,硕士,高级工程师,主要研究方向为智能用电及需求响应; 陈振宇(1973),男,研究员级高级工程师,主要研究方向为电力需求侧与能源管理; 杨阳(1994),女,硕士研究生,主要研究方向为负荷优化调度; 张凯恒(1995),男,硕士研究生,主要研究方向为分散式负荷响应及其不确定性分析; 张卫国(1980),男,硕士,高级工程师,主要研究方向为智能用电及需求响应; 朱庆(1981),男,博士,高级工程师,主要研究方向为电力系统自动化; 徐青山(1979),男,博士,教授,主要研究方向为智能配电网运行与规划。
  • 基金资助:
    国家电网公司总部科技项目(SGJS0000YXJS1700314)

Research on Load Demand Response Decision Considering Resource Correlation and Uncertainty

SONG Jie1,CHEN Zhenyu2,YANG Yang3,ZHANG Kaiheng3,ZHANG Weiguo1, ZHU Qing1,XU Qingshan3   

  1. 1.NARI Group Corporation, Nanjing 211106, China;2. State Grid Jiangsu Electric Power Co., Ltd., Nanjing 210029, China;3. Southeast University, Nanjing 211189, China
  • Online:2019-06-01
  • Supported by:
    This work is supported by State Grid Corporation of China Research Program (No. SGJS0000YXJS1700314).

摘要: 需求响应(demand response,DR)具有一定的不确定性,为了真实地反映各不确定性用户的响应情况,文章考虑了以空调为代表的不确定性用户资源之间在空间和时间上响应度的相关性。首先采用拉丁超立方采样法对响应度进行规模采样和相关性控制,得到反映响应相关性的样本。其次引入确定性资源,建立负荷聚合商(load aggregator,LA)的最优决策模型,并通过随机模拟和智能算法求解利润最高时的决策方案。最后通过算例分析了确定性资源、相关性及响应度的分布参数对决策方案和利润的影响,验证了订购确定性资源,签约负相关的不确定性用户,提高响应度的均值,降低响应度的方差均对利润有着显著的提高。

关键词: 需求响应(DR), 相关性, 不确定性, 拉丁超立方采样法, 负荷聚合商(LA)

Abstract: The demand response(DR)has certain uncertainty. In order to truly reflect the response of each uncertain user, this paper considers the correlation between spatial and temporal responsiveness of uncertain user resources represented by air conditioning. Firstly, the improved Latin hypercube sampling method is used to scale and correlate the responsivity, and the samples reflecting the response correlation are obtained. Secondly, the deterministic resources are introduced to establish the optimal decision model of the load aggregator(LA), and the decision-making scheme is solved by  the algorithm  when the profit is the highest through random simulation and intelligence. Finally, an example is given to analyze the influence of the distribution parameters of deterministic resources, correlation and responsiveness on decision-making schemes and profits. Measures that can increase the profit include ordering deterministic resources, signing negative related uncertain users, increasing the average of responsiveness, reducing the variance of responsiveness.

Key words: demand response(DR), correlation, uncertainty, Latin hypercube sampling, load aggregator(LA)

中图分类号: