工业用户连续参与需求响应的用户基线负荷精准计算方法

doi:10.11930/j.issn.1004-9649.202312036

中国电力 ›› 2024, Vol. 57 ›› Issue (3): 34-42.DOI: 10.11930/j.issn.1004-9649.202312036

• 数字化技术驱动的新型配电网 • 上一篇下一篇

工业用户连续参与需求响应的用户基线负荷精准计算方法

梁珩¹(), 黄耕², 侯宾², 杨玺³, 罗小虎³(), 张达¹()

1. 清华大学能源环境经济研究所，北京　100084
2. 北京邮电大学电子工程学院，北京　100876
3. 清华四川能源互联网研究院，四川成都　610000

收稿日期:2023-12-11 出版日期:2024-03-28 发布日期:2024-03-26
作者简介:梁珩（1994—），男，博士研究生，从事电力系统低碳转型、需求响应、电力系统规划和运行等研究，E-mail：liang-h19@mails.tsinghua.edu.cn
罗小虎（1987—），男，博士，从事电能负荷数字化管理平台、电力需求响应、城市能源电力碳排放监测、双碳相关信息化管理技术等研究，E-mail：luoxiaohu@tsinghua-eiri.org
张达（1987—），男，通信作者，博士，副教授，从事能源与应对气候变化的政策分析、综合评估模型和电力系统建模等研究，E-mail：zhangda@tsinghua.edu.cn
基金资助:
国家自然科学基金资助项目（71974109，72140005）。

Accurate Estimation Method of Customer Baseline Load for Continuous Participation of Industrial Users in Demand Response

Heng LIANG¹(), Geng HUANG², Bin HOU², Xi YANG³, Xiaohu LUO³(), Da ZHANG¹()

1. Institute of Energy Environment and Economy, Tsinghua University, Beijing 100084, China
2. School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China
3. Sichuan Energy Internet Research Institute, Tsinghua University, Chengdu 610000, China

Received:2023-12-11 Online:2024-03-28 Published:2024-03-26
Supported by:
This work is supported by National Natural Science Foundation of China (No.71974109, No.72140005).

摘要/Abstract

摘要：

提出了一种将K-means聚类分析与长短期记忆神经网络算法结合，通过工业同源组信息进行迁移学习优化的计算方法。该方法实现了对长时间连续参与需求响应的工业用户基线负荷的精准计算，提高了工业用户需求响应效果评价的准确性。通过城市级虚拟电厂平台采集的参与需求响应实践的工业用户电力负荷数据，验证了该方法的有效性。

关键词: 需求响应, 用户基线负荷计算, 聚类分析

Abstract:

A computational method combining K-means cluster analysis with long- and short-term memory neural network algorithm is proposed, and transfer learning is carried out by industrial homogeneous group information to further optimize the estimation effect. Accurate estimation of industrial customer power baseline load under long-term continuous response is realized, and the accuracy of the demand response effect evaluation of industrial customers is improved. The effectiveness of the method is verified by the load data of industrial customers participating in demand response practice collected by the city-level virtual power plant platform.

Key words: demand response, customer baseline load estimation, cluster analysis

梁珩, 黄耕, 侯宾, 杨玺, 罗小虎, 张达. 工业用户连续参与需求响应的用户基线负荷精准计算方法[J]. 中国电力, 2024, 57(3): 34-42.

Heng LIANG, Geng HUANG, Bin HOU, Xi YANG, Xiaohu LUO, Da ZHANG. Accurate Estimation Method of Customer Baseline Load for Continuous Participation of Industrial Users in Demand Response[J]. Electric Power, 2024, 57(3): 34-42.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202312036

https://www.electricpower.com.cn/CN/Y2024/V57/I3/34

图/表 16

图 1 CBL计算模型的整体框架

Fig.1 The overall framework of the CBL estimation model

图 2 典型用电特征提取流程

Fig.2 Typical power usage feature extraction process

图 3 资源分组流程

Fig.3 The resource grouping process

图 4 处理前后某企业5日负荷数据

Fig.4 5-day load data of an enterprise before and after treatment

图 5 聚类中心对比

Fig.5 Comparison of clustering centers

表 1 企业A的同组负荷曲线相似度

Table 1 Similarity of load curves within the same group in enterprise A

企业	S	企业	S
A	1.000	D	0.799
B	0.849	E	0.795
C	0.810	F	0.772

表 2 企业A CBL计算的迁移学习优化效果

Table 2 Optimization effect of CBL estimation for enterprise A by transfer learning

组内企业	MAE/kW	组内企业	MAE/kW
A	2 285.5	ABCD	1 540.8
AB	1 620.7	ABCDE	1 402.6
ABC	1 561.2	ABCDEF	1 470.3

图 6 迁移学习对企业A的CBL计算的优化效果

Fig.6 Optimization effect of transfer learning on CBL estimation of enterprise A

表 3 算例设计

Table 3 Case design

算例	响应日期	响应时段	单次响应容量/%
1	11月14日	12:00—20:00	10~30
2	11月14—18日	12:00—20:00	10~30

图 7 算例1下基于方案1~3得到的CBL

Fig.7 CBL based on 3 averaging methods in case 1

图 8 算例1下基于方案3~7得到的CBL

Fig.8 CBL based on method 3~7 in case 1

表 4 算例1下基于7种方案得到的CBL与真实值的偏差

Table 4 Deviation of CBL from the true value based on 7 methods in case 1

方案	MAE/kW	RMSE/kW	MAPE/%
1	4 405.76	4 745.28	15.59
2	4 921.30	5 106.05	17.36
3	4 353.98	4 556.30	15.36
4	2 320.26	2 950.66	8.03
5	3 608.09	3 727.07	12.78
6	1 269.38	1 455.03	4.50
7	708.75	753.31	2.49

图 9 算例2下基于3种平均法得到的CBL

Fig.9 CBL based on 3 averaging methods in case 2

图 10 算例2下基于方法3~7得到的CBL

Fig.10 CBL based on method 3~7 in case 2

表 5 算例2下基于7种方法得到的CBL与真实值的偏差

Table 5 Deviation of CBL from the true value based on 7 methods in case 2

方案	MAE/kW	RMSE/kW	MAPE/%
1	6670.74	6934.52	28.32
2	7164.99	7328.42	30.07
3	3696.04	3999.12	15.74
4	4195.92	4541.79	16.86
5	6224.07	6400.32	27.28
6	1380.38	1747.09	5.63
7	616.92	746.38	2.34

表 6 某企业的DR评估结果和激励金额

Table 6 DR contribution and incentive of the case enterprise

算例	方案	平均负荷响应率/%	平均激励金额/ (元·(kW·次)^–1)
1	参考值	100	12.00~12.00
	1	204.40~134.80	14.40~14.40
	2	216.60~138.90	14.40~14.40
	3	203.10~134.40	14.40~14.40
	4	48.40~82.80	0.00~9.94
	5	185.50~128.50	14.40~14.40
	6	107.00~102.30	12.84~12.28
	7	88.70~96.20	10.65~11.55
2	参考值	100	12.00~12.00
	1	257.20~152.40	14.40~14.17
	2	269.50~156.50	14.40~14.27
	3	184.40~128.10	14.21~13.88
	4	149.40~116.50	8.64~10.63
	5	247.40~149.10	14.40~14.21
	6	122.90~107.60	14.04~12.92
	7	90.00~96.70	10.80~11.60

参考文献 26

1	蒋棹骏, 向月, 谈竹奎, 等. 计及需求响应的高比例清洁能源园区储能容量优化配置[J]. 中国电力, 2023, 56 (12): 147- 155, 163.
	JIANG Zhaojun, XIANG Yue, TAN Zhukui, et al. Optimal allocation of energy storage capacity in high proportion clean energy parks considering demand response[J]. Electric Power, 2023, 56 (12): 147- 155, 163.
2	黄蔚亮, 苏志鹏, 梁欣怡, 等. 考虑可调市场和外部需求响应的虚拟电厂优化运行策略[J]. 中国电力, 2023, 56 (12): 156- 163.
	HUANG Weiliang, SU Zhipeng, LIANG Xinyi, et al. Optimal operation strategy for virtual power plant considering regulation market and external demand response[J]. Electric Power, 2023, 56 (12): 156- 163.
3	毕锐, 王孝淦, 袁华凯, 等. 考虑供需双侧响应和碳交易的氢能综合能源系统鲁棒调度[J]. 电力系统保护与控制, 2023, 51 (12): 122- 132.
	BI Rui, WANG Xiaogan, YUAN Huakai, et al. Robust dispatch of a hydrogen integrated energy system considering double side response and carbon trading mechanism[J]. Power System Protection and Control, 2023, 51 (12): 122- 132.
4	赵梓潼, 顾兵. 需求响应下基于电动汽车负荷聚合商的充放电电价与时段研究[J]. 东北电力大学学报, 2023, 43 (6): 79- 86.
	ZHAO Zitong, GU Bing. Research on charging and discharging price and time period based on electric vehicle load aggregator under demand response[J]. Journal of Northeast Electric Power University, 2023, 43 (6): 79- 86.
5	申洪涛, 刘宝铭, 任鹏, 等. 基于辨识解耦的含高渗透分布式光伏用户集群基线负荷估计方法[J]. 电力系统保护与控制, 2022, 50 (3): 164- 173.
	SHEN Hongtao, LIU Baoming, REN Peng, et al. Aggregated baseline load estimation method under high distributed photovoltaic penetration based on identification and decoupling[J]. Power System Protection and Control, 2022, 50 (3): 164- 173.
6	刘金朋, 刘胡诗涵, 张雨菲, 等. 考虑居民用户可调节潜力的负荷聚合商日前投标决策优化模型研究[J]. 智慧电力, 2024, (2): 71- 78, 122.
	LIU Jinpeng, LIU Hushihan, ZHANG Yufei, et al. Day-ahead bidding decision optimization model of load aggregators considering adjustable potential of residential users[J]. Smart Power, 2024, (2): 71- 78, 122.
7	陈海鹏, 周越豪, 赵畅, 等. 考虑高载能负荷参与的多时间尺度风电消纳调度[J]. 东北电力大学学报, 2022, 42 (6): 39- 51.
	CHEN Haipeng, ZHOU Yuehao, ZHAO Chang, et al. Multi-time scale wind power consumption scheduling with high energy load participation is considered[J]. Journal of Northeast Electric Power University, 2022, 42 (6): 39- 51.
8	李东东, 张凯, 姚寅, 等. 基于信息间隙决策理论的电动汽车聚合商日前需求响应调度策略[J]. 电力系统保护与控制, 2022, 50 (24): 101- 111.
	LI Dongdong, ZHANG Kai, YAO Yin, et al. Day-ahead demand response scheduling strategy of an electric vehicle aggregator based on information gap decision theory[J]. Power System Protection and Control, 2022, 50 (24): 101- 111.
9	丁琦欣, 覃洪培, 万灿, 等. 基于机会约束规划的配电网分布式光伏承载能力评估[J]. 东北电力大学学报, 2022, 42 (6): 28- 38.
	DING Qixin, QIN Hongpei, WAN Can, et al. Chance-constrained optimization-based distributed photovoltaic hosting capacity assessment of distribution networks[J]. Journal of Northeast Electric Power University, 2022, 42 (6): 28- 38.
10	刘雪飞, 刘洋, 马国真, 等. 考虑负荷差异化需求响应的配电网多目标扩展规划[J]. 电力系统保护与控制, 2022, 50 (22): 131- 141.
	LIU Xuefei, LIU Yang, MA Guozhen, et al. Multi-objective extended planning for a distribution network considering demarcation of demand response schemes[J]. Power System Protection and Control, 2022, 50 (22): 131- 141.
11	何胜, 徐玉婷, 陈宋宋, 等. 我国电力需求响应发展成效及“十四五” 工作展望[J]. 电力需求侧管理, 2021, 23 (6): 1- 6.
	HE Sheng, XU Yuting, CHEN Songsong, et al. Prospect and the 14th Five-Year Plan of power demand response development effect in China[J]. Power Demand Side Management, 2021, 23 (6): 1- 6.
12	贾巍, 黄裕春. 基于小样本数据差分扩容的微电网负荷预测方法[J]. 中国电力, 2023, 56 (8): 151- 156, 165.
	JIA Wei, HUANG Yuchun. Method of load forecasting in microgrid based on differential expansion of small sample data[J]. Electric Power, 2023, 56 (8): 151- 156, 165.
13	欧阳福莲, 王俊, 周杭霞. 基于改进迁移学习和多尺度CNN-BiLSTM-Attention的短期电力负荷预测方法[J]. 电力系统保护与控制, 2023, 51 (2): 132- 140.
	OUYANG Fulian, WANG Jun, ZHOU Hangxia. Short-term power load forecasting method based on improved hierarchical transfer learning and multi-scale CNN-BiLSTM-Attention[J]. Power System Protection and Control, 2023, 51 (2): 132- 140.
14	李文武, 石强, 李丹, 等. 基于VMD和PSO-SVR的短期电力负荷多阶段优化预测[J]. 中国电力, 2022, 55 (8): 171- 177.
	LI Wenwu, SHI Qiang, LI Dan, et al. Multi-stage optimization forecast of short-term power load based on VMD and PSO-SVR[J]. Electric Power, 2022, 55 (8): 171- 177.
15	LEE E, LEE K, LEE H, et al. Defining virtual control group to improve customer baseline load calculation of residential demand response[J]. Applied Energy, 2019, 250, 946- 958. DOI
16	杨彪, 颜伟, 莫静山. 考虑源荷功率随机性和相关性的主导节点选择与无功分区方法[J]. 电力系统自动化, 2021, 45 (11): 61- 67.
	YANG Biao, YAN Wei, MO Jingshan. Pilot-bus selection and network partitioning method considering randomness and correlation of source-load power[J]. Automation of Electric Power Systems, 2021, 45 (11): 61- 67.
17	高欣, 纪维佳, 赵兵, 等. 不平衡数据集下基于CVAE-CNN模型的智能电表故障多分类方法[J]. 电网技术, 2021, 45 (8): 3052- 3060.
	GAO Xin, JI Weijia, ZHAO Bing, et al. Multi-classification method of smart meter fault types based on CVAE-CNN model under imbalanced dataset[J]. Power System Technology, 2021, 45 (8): 3052- 3060.
18	黄南天, 郭玉, 赵暄远. 计及辐照区间划分的含光伏电源配电网源-荷联合场景生成[J]. 东北电力大学学报, 2023, 43 (5): 78- 84.
	HUANG Nantian, GUO Yu, ZHAO Xuanyuan. Combined source-load scenario generation for PV-containing distribution networks with calculation and irradiation interval classification[J]. Journal of Northeast Electric Power University, 2023, 43 (5): 78- 84.
19	肖白, 李梦雪, 尉博旭. 基于AP聚类-跳转持续MC的风电出力时间序列模拟生成方法研究[J]. 东北电力大学学报, 2023, 43 (1): 35- 44.
	XIAO Bai, LI Mengxue, YU Boxu. Research on simulation method to generate wind power output time series based on AP clustering-transition and persistence Markov chain[J]. Journal of Northeast Electric Power University, 2023, 43 (1): 35- 44.
20	张露, 颜宏文, 马瑞. 基于改进DBSCAN-RNN的电力负荷建模及可调特征提取[J]. 智慧电力, 2023, 51 (3): 39- 45.
	ZHANG Lu, YAN Hongwen, MA Rui. Power load modeling and adjustable feature extraction based on improved DBSCAN-RNN[J]. Smart Power, 2023, 51 (3): 39- 45.
21	RENCHER A C, CHRISTENSEN W F. Classification analysis: allocation of observations to groups[J]. Methods of Multivariate Analysis, 2003: 299–321.
22	邝梓佳, 邱桂华, 汤志锐. 基于迁移学习的OMS配网瞬态稳定性控制方法[J]. 信息技术, 2024, 48 (1): 182- 188.
	KUANG Zijia, QIU Guihua, TANG Zhirui. Transient stability control method of OMS distribution network based on Transfer Learning[J]. Information Technology, 2024, 48 (1): 182- 188.
23	ZHU Z D, LIN K X, JAIN A K, et al. Transfer learning in deep reinforcement learning: a survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45 (11): 13344- 13362.
24	赵源上, 林伟芳. 基于皮尔逊相关系数融合密度峰值和熵权法典型场景研究[J]. 中国电力, 2023, 56 (5): 193- 202.
	ZHAO Yuanshang, LIN Weifang. Research on typical scenarios based on fusion density peak value and entropy weight method of pearson's correlation coefficient[J]. Electric Power, 2023, 56 (5): 193- 202.
25	安军, 周永超, 周毅博, 等. 考虑风荷不确定性的电源无功电压调差系数整定方法研究[J]. 东北电力大学学报, 2023, 43 (4): 30- 38.
	AN Jun, ZHOU Yongchao, ZHOU Yibo, et al. Optimization method for var-voltage adjustment CoefficientConsidering wind power and load uncertainty[J]. Journal of Northeast Electric Power University, 2023, 43 (4): 30- 38.
26	刘文丽, 张涛, 杨晓雷, 等. 计及负荷随机性含风电电力系统TCSC多目标优化配置[J]. 电力系统保护与控制, 2023, 51 (5): 58- 69.
	LIU Wenli, ZHANG Tao, YANG Xiaolei, et al. Multi-objective optimal allocation of TCSC for a power system for wind power and load randomness[J]. Power System Protection and Control, 2023, 51 (5): 58- 69.

[1]	许文俊, 马刚, 姚云婷, 孟宇翔, 李伟康. 考虑绿证-碳交易机制与混氢天然气的工业园区多能优化调度[J]. 中国电力, 2025, 58(2): 154-163.
[2]	谭玲玲, 汤伟, 楚冬青, 李竞锐, 张玉敏, 吉兴全. 基于主从博弈的电热氢综合能源系统优化运行[J]. 中国电力, 2024, 57(9): 136-145.
[3]	徐峰亮, 王克谦, 王文豪, 王鹏, 王文烨, 张帅, 赵凤展. 计及激励型需求响应的低压配电网混合储能优化配置[J]. 中国电力, 2024, 57(6): 90-101.
[4]	张彩玲, 王爽, 葛淑娜, 潘登, 张岩, 韩伟, 段文岩. 计及灵活需求响应和碳-绿证交易的综合能源系统优化调度[J]. 中国电力, 2024, 57(5): 14-25.
[5]	李方姝, 余昆, 陈星莺, 华昊辰. 碳约束下基于双重博弈的电力零售商售电价格决策优化[J]. 中国电力, 2024, 57(5): 126-136.
[6]	谭玲玲, 汤伟, 楚冬青, 于子涵, 吉兴全, 张玉敏. 考虑电-氢一体化的微电网低碳-经济协同优化调度[J]. 中国电力, 2024, 57(5): 137-148.
[7]	高月芬, 员成博, 孔凡鹏, 王雪松. 需求响应激励下耦合电转气、碳捕集设备的综合能源系统优化[J]. 中国电力, 2024, 57(4): 32-41.
[8]	甘润东, 龙玉江, 汤杰, 何熙, 罗鸿轩, 金鑫. 计及负荷时空转移需求响应的数据中心聚合商最优运行策略[J]. 中国电力, 2024, 57(3): 20-26.
[9]	高志远, 庄卫金, 耿建, 李峰, 薛必克, 杨晓雷, 白柯鞠. 基于经济人假设的负荷侧资源市场化调节作用机理分析[J]. 中国电力, 2024, 57(3): 213-223.
[10]	唐志辉, 张欣, 宁艺飞, 黄璜, 余昆, 华昊辰, 曹佳伟, 郑云天. 基于联盟链技术的商业建筑转供电主体需求响应策略[J]. 中国电力, 2024, 57(12): 109-119.
[11]	张晓萱, 薛松, 许野, 许轶, 丁泽宇, 孙庆凯. 考虑可调节负荷减碳降碳价值的需求响应运行决策模型[J]. 中国电力, 2024, 57(11): 151-160.
[12]	赵先海, 刘晓峰, 季振亚, 李峰, 刘国宝. 考虑居民用户动态行为的负荷聚合商决策分析[J]. 中国电力, 2024, 57(10): 179-189.
[13]	苏湘波, 吕睿可, 郭鸿业, 陈启鑫. 基于负荷台阶的工业需求响应用户优选方法[J]. 中国电力, 2024, 57(1): 18-29.
[14]	王世杰, 冯天波, 孙宁, 何可, 李嘉文, 杨程, 崔昊杨. 考虑电-气-热耦合和需求响应的虚拟电厂优化调度策略[J]. 中国电力, 2024, 57(1): 101-114.
[15]	吴桐, 惠红勋, 张洪财. 商业建筑空调系统参与城市电网负荷调控综述[J]. 中国电力, 2023, 56(7): 1-11.

工业用户连续参与需求响应的用户基线负荷精准计算方法

Accurate Estimation Method of Customer Baseline Load for Continuous Participation of Industrial Users in Demand Response

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 26

相关文章 15

编辑推荐

Metrics

模态框（Modal）标题

工业用户连续参与需求响应的用户基线负荷精准计算方法

Accurate Estimation Method of Customer Baseline Load for Continuous Participation of Industrial Users in Demand Response

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 26

相关文章 15

编辑推荐

Metrics