刘继东;韩学山;韩伟吉;张利;

• 1:电网智能化调度与控制教育部重点实验室(山东大学)

摘要(Abstract)：

为满足需求响应机制中描述用户行为规律的需要,提出一种电力用户需求响应行为的模型与算法。在获取足够的用户历史数据的基础上,通过支持向量机(support vector machine,SVM)回归进行数据挖掘,建立了电力用户在分时电价下的响应行为模型。该方法以用户响应的影响因素分析为基础,确定了回归模型的输入与输出属性;并通过定义等效电价比率,构建了含丰富数据信息的训练样本;最后采用网格搜索法选择SVM回归的最佳参数,实现了回归模型的高精度预测。该模型实现了电力用户在分时电价下行为规律的模拟,可揭示用户响应电量变化与分时电价政策激励力度间的关系,从而为更多研究提供基础数据。仿真分析证明了该模型和算法的有效性和合理性。

关键词(KeyWords)： 需求响应;分时电价;用户行为;支持向量机;回归分析

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(51077087);; 山东省自然科学基金项目(Y2008F19)~~

作者(Author): 刘继东;韩学山;韩伟吉;张利;

Email:

DOI: 10.13335/j.1000-3673.pst.2013.10.030

参考文献(References)：

• [1]张钦,王锡凡,付敏,等.需求响应视角下的智能电网[J].电力系统自动化,2009,33(17):49-55.Zhang Qin,Wang Xifan,Fu Min,et al.Smart grid from theperspective of demand response[J].Automation of Electric PowerSystems,2009,33(17):49-55(in Chinese).
• [2]李来福,柳焯.需求侧管理在缓解紧急态势中的作用[J].电网技术,2008,32(7):56-60.Li Laifu,Liu Zhuo.Role of demand side management played inalleviation of urgent situation[J].Power System Technology,2008,32(7):56-60(in Chinese).
• [3]赵鸿图,朱治中,于尔铿.电力市场中需求响应市场与需求响应项目研究[J].电网技术,2010,34(5):146-153.Zhao Hongtu,Zhu Zhizhong,Yu Erkeng.Study on demand responsemarkets and programs in electricity markets[J].Power SystemTechnology,2010,34(5):146-153(in Chinese).
• [4]Federal Energy Regulatory Commission.A national assessment ofdemand response potential[EB/OL].[2012-12-10].http://www.ferc.gov/industries/electric/indus-act/demand-response/dr-potential/assessment.asp.
• [5]Woychik E C.Optimizing demand response[J].Public UtilitiesFortnightly,2008(5):52-56.
• [6]North American Electric Reliability Corporation.Potential reliabilityimpact of emerging flexible recourses,IVGTF task-1-5[R].Princeton:NERC,2011.
• [7]Kirschen D S.Demand-side view of electricity markets[J].IEEETransactions on Power Systems,2003,18(2):520-527.
• [8]辛洁晴,程浩忠.电力需求短期分时价格弹性的特性及其应用[J].电力系统自动化,2007,31(10):32-35.Xin Jieqing,Cheng Haozhong.Properties of short-term TOU priceelasticities of electricity demand and their application[J].Automationof Electric Power Systems,2007,31(10):32-35(in Chinese).
• [9]刘严,谭忠富,乞建勋.峰谷分时电价设计的优化模型[J].中国管理科学,2005,13(5):87-92.Liu Yan,Tan Zhongfu,Qi Jianxun.Research on TOU rate designoptimization model[J].Chinese Journal of Management Science,2005,13(5):87-92(in Chinese).
• [10]阮文骏,王蓓蓓,李扬,等.峰谷分时电价下的用户响应行为研究[J].电网技术,2012,36(7):86-93.Luan Wenjun,Wang Beibei,Li Yang,et al.Customer responsebehavior in time-of-use price[J].Power System Technology,2012,36(7):86-93(in Chinese).
• [11]袁家海,李蒙,胡兆光.基于多智能体模拟方法的大用户峰谷电价响应研究[J].电网技术,2005,29(23):44-49.Yuan Jiahai,Li Meng,Hu Zhaoguang.Research on large customerresponse under time-of-use electricity pricing policy based onmulti-agent system simulation[J].Power System Technology,2005,29(23):44-49(in Chinese).
• [12]谈金晶,王蓓蓓,李扬.基于多智能体的用户分时电价响应模型[J].电网技术,2012,36(2):257-263.Tan Jinjing,Wang Beibei,Li Yang.Modeling of user response totime-of-use price based on multi-agent technology[J].Power SystemTechnology,2012,36(2):257-263(in Chinese).
• [13]Celebie E,Fuller J D.A model for efficient consumer pricing schemesin electricity markets[J].IEEE Trans on Power Systems,2007,22(1):60-67.
• [14]姚珺玉,刘俊勇,刘友波,等.计及时滞指标灵敏度的用户电价响应模式划分方法[J].电网技术,2010,34(4):30-36.Yao Junyu,Liu Junyong,Liu Youbo,et al.An approach to dividecustomer price response modes taking comprehensive sensitivity ofthe time delay index into account[J].Power System Technology,2010,34(4):30-36(in Chinese).
• [15]Roos J G,Kern C F.Modeling customer demand response to dynamicprice signals using artificial intelligence[C]//8th InternationalConference on Metering and Tariffs for Energy Supply.Brighton:IEEE,1996:213-217.
• [16]何永秀,王冰,熊威,等.基于模糊综合评价的居民智能用电行为分析与互动机制设计[J].电网技术,2012,36(10):247-258.He Yongxiu,Wang Bing,Xiong Wei,et al.Analysis of residents’smart electricity consumption behavior based on fuzzy syntheticevaluation and the design of interactive mechanism[J].Power SystemTechnology,2012,36(10):247-258(in Chinese).
• [17]Mathieu J L,Gadgil A J,Callaway D S,et al.Characterizing theresponse of commercial and industrial facilities to dynamic pricingsignals from the utility[EB/OL].[2012-12-10].http://escholarship.org/uc/item/2hh094jx.
• [18]Cortes C,Vapnik V.Support-vector networks[J].Machine learning,1995,20(3):273-297.
• [19]潘峰,程浩忠,杨镜非,等.基于支持向量机的电力系统短期负荷预测[J].电网技术,2004,28(21):39-42.Pan Feng,Cheng Haozhong,Yang Jingfei,et al.Power systemshort-term load forecasting based on support vector machines[J].Power System Technology,2004,28(21):39-42(in Chinese).
• [20]李瑾,刘金朋,王建军.采用支持向量机和模拟退火算法的中长期负荷预测方法[J].中国电机工程学报,2011,31(16):63-66.Li Jin,Liu Jinpeng,Wang Jianjun.Mid-long term load forecastingbased on simulated annealing and SVM algorithm[J].Proceedings ofthe CSEE,2011,31(16):63-66(in Chinese).
• [21]李元诚,刘克文.面向大规模样本的核心向量回归电力负荷快速预测方法[J].中国电机工程学报,2010,30(28):33-38.Li Yuancheng,Liu Kewen.A rapid electric load forecasting methodusing core vector regression for large scale data set[J].Proceedings ofthe CSEE,2010,30(28):33-38(in Chinese).
• [22]刘运龙,刘继东,殷树刚.山东电力营销技术支持系统的设计与实现[J].中国电力,2008,14(3):75-80.Liu Yunlong,Liu Jidong,Yin Shugang.Design and implementationof Shandong electric power marketing technology supporting system[J].Electric Power,2008,14(3):75-80(in Chinese).
• [23]谢宏,魏江平,刘鹤立.短期负荷预测中支持向量机模型的参数选取和优化方法[J].中国电机工程学报,2006,26(22):17-22.Xie Hong,Wei Jiangping,Liu Heli.Parameter selection andoptimization method of SVM model for short-term load forecasting[J].Proceedings of the CSEE,2006,26(22):17-22(in Chinese).
• [24]闫国华,朱永生.支持向量机回归的参数选择方法[J].计算机工程,2009,35(13):218-220.Yan Guohua,Zhu Yongsheng.Parameters selection method forsupport vector machine regression[J].Computer Engineering,2009,35(13):218-220(in Chinese).