基于k-shape聚类和TCN-Attention-XGBoost的基线负荷估计方法

沈杰; 邢海军; 俞钱; 施怡沁

doi:10.19912/j.0254-0096.tynxb.2024-0661

PDF(2003 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (8) : 676-685. DOI: 10.19912/j.0254-0096.tynxb.2024-0661

基于k-shape聚类和TCN-Attention-XGBoost的基线负荷估计方法

沈杰, 邢海军, 俞钱, 施怡沁

作者信息 +

BASELINE LOAD ESTIMATION METHOD BASED ON K-SHAPE CLUSTERING AND TCN-ATTENTION-XGBOOST

Shen Jie, Xing Haijun, Yu Qian, Shi Yiqin

Author information +

文章历史 +

摘要

针对传统基线负荷估计方法在面对多变的电力负荷消耗时准确度不足的挑战,提出一种结合k-shape聚类算法和基于时间卷积网络（TCN）、注意力机制和极端梯度提升（XGBoost）的估计方法。首先利用TCN-Attention模型进行时间序列预测,提取负荷的时间特征;然后采用k-shape聚类算法对用户负荷数据进行聚类,以找出在形状和趋势上相似度高的负荷序列,并采用XGBoost算法捕捉负荷的空间特征。最后,将两者的估计值进行误差倒数法组合得到最终的估计结果。利用实际算例进行验证,并将所提方法与其他方法进行对比,结果表明了所提方法的有效性。

Abstract

To address the challenge of insufficient accuracy in traditional baseline load estimation methods under dynamic electricity consumption patterns, this study proposes an integrated approach combining the k-shape clustering algorithm with a hybrid model based on temporal convolutional network （TCN）, attention mechanism, and extreme gradient boosting （XGBoost）. Firstly, the TCN-Attention model is employed for time series forecasting to extract temporal features of load profiles. Subsequently, the k-shape clustering algorithm is applied to group user load data into clusters with high similarity in shape and trend, while the XGBoost algorithm is utilized to capture spatial characteristics of load patterns. Finally, the individual estimates from both components are integrated using an inverse error weighting method to generate the final load estimation. The proposed method is validated through real-world case studies and compared other methods. The results demonstrate the effectiveness of the proposed method.

导出引用

沈杰, 邢海军, 俞钱, 施怡沁. 基于k-shape聚类和TCN-Attention-XGBoost的基线负荷估计方法[J]. 太阳能学报. 2025, 46(8): 676-685 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0661

Shen Jie, Xing Haijun, Yu Qian, Shi Yiqin. BASELINE LOAD ESTIMATION METHOD BASED ON K-SHAPE CLUSTERING AND TCN-ATTENTION-XGBOOST[J]. Acta Energiae Solaris Sinica. 2025, 46(8): 676-685 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0661

中图分类号： TM714

参考文献

[1] 国家能源局. 国家能源局发布2023年全国电力工业统计数据[EB/OL]. (2024-01-26)[2024-01-26]. http://www.nea.gov.cn/2024-01/26/c_1310762246.htm.
National Energy Administration. National Energy Administration releases2023 national electric power industry statistics[EB/OL]. (2024-01-26)[2024-01-26]. http://www.nea.gov.cn/2024-01/26/c_1310762246.htm.
[2] 朱凌志, 陈宁, 韩华玲. 风电消纳关键问题及应对措施分析[J]. 电力系统自动化, 2011, 35(22): 29-34.
ZHU L Z, CHEN N, HAN H L.Key problems and solutions of wind power accommodation[J]. Automation of electric power systems, 2011, 35(22): 29-34.
[3] 舒印彪, 张智刚, 郭剑波, 等. 新能源消纳关键因素分析及解决措施研究[J]. 中国电机工程学报, 2017, 37(1): 1-9.
SHU Y B, ZHANG Z G, GUO J B, et al.Study on key factors and solution of renewable energy accommodation[J]. Proceedings of the CSEE, 2017, 37(1): 1-9.
[4] 曾博, 杨雍琦, 段金辉, 等. 新能源电力系统中需求侧响应关键问题及未来研究展望[J]. 电力系统自动化, 2015, 39(17): 10-18.
ZENG B, YANG Y Q, DUAN J H, et al.Key issues and research prospects for demand-side response in alternate electrical power systems with renewable energy sources[J]. Automation of electric power systems, 2015, 39(17): 10-18.
[5] 张舒菡. 智能电网条件下的需求响应关键技术[J]. 电子技术与软件工程, 2017(6): 219-220.
ZHANG S H.Key technologies of demand response under smart grid conditions[J]. Electronic technology & software engineering, 2017(6): 219-220.
[6] 范帅, 危怡涵, 何光宇, 等. 面向新型电力系统的需求响应机制探讨[J]. 电力系统自动化, 2022, 46(7): 1-12.
FAN S, WEI Y H, HE G Y, et al.Discussion on demand response mechanism for new power systems[J]. Automation of electric power systems, 2022, 46(7): 1-12.
[7] 牛文娟, 王磊, 李扬. 需求响应项目中用户基线负荷计算方法及其应用[J]. 东南大学学报(自然科学版), 2014, 44(3): 556-560.
NIU W J, WANG L, LI Y.Calculation method and application of customer baseline load in demand response project[J]. Journal of Southeast University (natural science edition), 2014, 44(3): 556-560.
[8] 胡可心, 李康平, 刘春阳, 等. 动态分时电价下居民用户需求响应基线负荷预测方法[J/OL]. 电测与仪表, 2023: 1-11. (2023-01-17). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CJFD&dbname=CJFD&filename=DCYQ20230116002.
HU K X, LI K P, LIU C Y, et al. A baseline load forecasting method for residential demand response under dynamic time-of-use electricity price[J/OL]. Electrical measurement & instrumentation, 2023: 1-11. (2023-01-17). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CJFD&dbname=CJFD&filename=DCYQ20230116002.
[9] 刘春阳, 李康平, 纪陵, 等. 基于聚类-估计联动的需求响应集群基线负荷估计方法[J]. 电力系统自动化, 2023, 47(2): 79-87.
LIU C Y, LI K P, JI L, et al.Clustering-estimation linkage based estimation method for aggregated baseline loads of demand response[J]. Automation of electric power systems, 2023, 47(2): 79-87.
[10] COUGHLIN K, PIETTE M A, GOLDMAN C, et al.Estimating demand response load impacts: evaluation of baseline load models for non-residential buildings in California[J]. Office of scientific & technical information technical reports, 2008: 12-13.
[11] WIJAYA T K, VASIRANI M, ABERER K.When bias matters: an economic assessment of demand response baselines for residential customers[J]. IEEE transactions on smart grid, 2014, 5(4): 1755-1763.
[12] MOHAJERYAMI S, DOOSTAN M, ASADINEJAD A.An investigation of the relationship between accuracy of customer baseline calculation and efficiency of Peak Time Rebate program[C]//2016 IEEE Power and Energy Conference at Illinois (PECI). Urbana, IL, USA, 2016: 1-8.
[13] WI Y M, KIM J H, JOO S K, et al.Customer baseline load (CBL) calculation using exponential smoothing model with weather adjustment[C]//2009 Transmission & Distribution Conference & Exposition: Asia and Pacific. Seoul, Korea, 2009: 1-4.
[14] 王啸峰, 苏慧玲, 宋天立, 等. 基于负荷细分的差异化用户基线负荷预测[J]. 电力工程技术, 2018, 37(6): 33-38.
WANG X F, SU H L, SONG T L, et al.Differential customer baseline load forecasting based on load subdivision[J]. Electric power engineering technology, 2018, 37(6): 33-38.
[15] 刘顺桂, 杨佳驹, 王磊. 考虑用户用电特性的基线负荷计算方法[J]. 电力需求侧管理, 2016, 18(3): 17-22.
LIU S G, YANG J J, WANG L.Baseline load calculation with considering customer different electrical characteristics[J]. Power demand side management, 2016, 18(3): 17-22.
[16] 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 H T, LIU B M, REN P, 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.
[17] 付文杰, 王喻玺, 申洪涛, 等. 基于拉丁超立方抽样和场景消减的居民用户基线负荷估计方法[J]. 电网技术, 2022, 46(6): 2298-2307.
FU W J, WANG Y X, SHEN H T, et al.Residential customer baseline load estimation based on Latin hypercube sampling and scenario subtraction[J]. Power system technology, 2022, 46(6): 2298-2307.
[18] 马庆, 李歧强. 基于电力需求响应的公共建筑基线负荷预测[J]. 山东大学学报(工学版), 2011, 41(2): 114-118.
MA Q, LI Q Q.Public buildings baseline load forecasting based on demand response in electric power[J]. Journal of Shandong University (engineering science), 2011, 41(2): 114-118.
[19] PRIOLKAR J, SREERAJ E S, THAKUR A.Analysis of consumer baseline for demand response implementation: a case study[C]//2020 7th International Conference on Signal Processing and Integrated Networks (SPIN). Noida, India, 2020: 89-94.
[20] SONG T L, LI Y, ZHANG X P, et al.A cluster-based baseline load calculation approach for individual industrial and commercial customer[J]. Energies, 2019, 12(1): 64.
[21] TAO P, XU F, DONG Z B, et al.Graph convolutional network-based aggregated demand response baseline load estimation[J]. Energy, 2022, 251: 123847.
[22] GE X X, XU F, WANG Y X, et al.Spatio-temporal two-dimensions data based customer baseline load estimation approach using LASSO regression[J]. IEEE transactions on industry applications, 2022, 58(3): 3112-3122.
[23] 吴珺玥, 赵二刚, 郭增良, 等. 基于Spearman系数和TCN的光伏出力超短期多步预测[J]. 太阳能学报, 2023, 44(9): 180-186.
WU J Y, ZHAO E G, GUO Z L, et al.Ultra-short-term photovoltaic power multi-step prediction based on Spearman coefficient and TCN[J]. Acta energiae solaris sinica, 2023, 44(9): 180-186.
[24] PAPARRIZOS J, GRAVANO L.K-shape: efficient and accurate clustering of time series[C]//Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data. Melbourne Victoria Australia, 2015: 1855-1870.
[25] 魏勇, 李学军, 李万伟, 等. 基于空间密度聚类和K-shape算法的城市综合体负荷模式聚类方法[J]. 电力系统保护与控制, 2021, 49(14): 37-44.
WEI Y, LI X J, LI W W, et al.Load pattern clustering method of an urban complex based on DBSCAN and K-shape algorithm[J]. Power system protection and control, 2021, 49(14): 37-44.
[26] 赵洋, 王瀚墨, 康丽, 等. 基于时间卷积网络的短期电力负荷预测[J]. 电工技术学报, 2022, 37(5): 1242-1251.
ZHAO Y, WANG H M, KANG L, et al.Temporal convolution network-based short-term electrical load forecasting[J]. Transactions of China Electrotechnical Society, 2022, 37(5): 1242-1251.
[27] 刘国海, 孙文卿, 吴振飞, 等. 基于Attention-GRU的短期光伏发电功率预测[J]. 太阳能学报, 2022, 43(2): 226-232.
LIU G H, SUN W Q, WU Z F, et al.Short-term photovoltaic power forecasting based on Attention-GRU model[J]. Acta energiae solaris sinica, 2022, 43(2): 226-232.
[28] 王献志, 曾四鸣, 周雪青, 等. 基于XGBoost联合模型的光伏发电功率预测[J]. 太阳能学报, 2022, 43(4): 236-242.
WANG X Z, ZENG S M, ZHOU X Q, et al.Power forecast of photovoltaic generation based on XGBoost combined model[J]. Acta energiae solaris sinica, 2022, 43(4): 236-242.
[29] 伍子东, 董萍. 基线计算偏差对负荷基线设计的影响研究[J]. 电网技术, 2021, 45(11): 4407-4415.
WU Z D, DONG P.Research on load baseline design based on calculation deviation[J]. Power system technology, 2021, 45(11): 4407-4415.
[30] Irish social science data archive. Commission for energy regulation(CER)smart metering project[EB/OL].[2024-2-26]. http://www.ucd.ie/issda/data/commissionforenergyregulationcer.
[31] 李姚旺, 张宁, 杜尔顺, 等. 基于碳排放流的电力系统低碳需求响应机制研究及效益分析[J]. 中国电机工程学报, 2022, 42(8): 2830-2842.
LI Y W, ZHANG N, DU E S, et al.Mechanism study and benefit analysis on power system low carbon demand response based on carbon emission flow[J]. Proceedings of the CSEE, 2022, 42(8): 2830-2842.
[32] 叶宇静, 邢海军, 米阳, 等. 考虑低碳需求响应及主从博弈的综合能源系统低碳优化调度[J]. 电力系统自动化, 2024, 48(9): 34-43.
YE Y J, XING H J, MI Y, et al.Low-carbon optimal dispatching of integrated energy system considering low-carbon demand response and Stackelberg game[J]. Automation of electric power systems, 2024, 48(9): 34-43.