Two-Stage Power User Classification Method Based on Digital Feature Portraits of Power Consumption Behavior

doi:10.12204/j.issn.1000-7229.2022.02.009

Abstract

Abstract:

Fine power consumption behavior portrait and classification has been one of the key factors for power enterprises to accurately grasp the electricity consumption law of power consumers, improve service level and market competitiveness. To solve the issues of fragmentary portraits of electricity consumption behavior in current power user classification research, base classifier redundancy and class imbalance in ensemble learning load classification, a two-stage power consumer classification method based on digital feature portraits of power consumption behavior is proposed. In the first stage, a classification method for power user daily load curves is proposed combining spectral clustering and integrated strong base classifier. Firstly, a strong base classifier is developed on the basis of LSTM network to improve the weak learning capability of base classifier in ensemble learning. Secondly, an Optimal Selection Ensemble (OSE) strategy based on minimum regularized surrogate empirical risk is proposed to solve the problem of base classifier redundancy. Thirdly, a Density Based Gaussian Synthetic (DBGS) minority over-sampling technique is proposed for class imbalance. In the second stage, the power consumption behavior portraits with daily load-pattern occurrence probability as the digital features are constructed, and the portraits are classified by spectral clustering. Finally, the effectiveness of the proposed method is verified by the measured user load data.

Key words: power user classification, digital feature portrait, load curve classification, class imbalance, optimization selection ensemble

CLC Number:

TM714

WANG Lei, LIU Yang, LI Wenfeng, ZHANG Jie, XU Lixiong, XING Zheming. Two-Stage Power User Classification Method Based on Digital Feature Portraits of Power Consumption Behavior[J]. ELECTRIC POWER CONSTRUCTION, 2022, 43(2): 70-80.

Figures/Tables 13

Fig.1 Schematic diagram of DBGS algorithm

Fig.2 Internal structure of LSTMP unit

Fig.3 ALN-LSTMP classification model

Fig.4 Schematic diagram of OSE-ALN-LSTMP

Fig.5 Selection of optimal number of clusters

Fig.6 Regulation characteristics of super parameter λ

Table 1 Comparison of classification accuracy of algorithms

Table 2 Comparison of classification accuracy of algorithms

Fig.7 Results of load curve classification

Fig.8 Seven typical load patterns

Fig.9 Eight typical user portraits

Fig.10 Comparison of user classification methods

Fig.11 Fig.A1 High dimensional visualization window of user classification results

References 23

[1]	李华强, 李旭翔, 阚力丰. 能源互联网背景下综合能源服务市场运营模式及关键技术[J]. 工程科学与技术, 2020, 52(4):13-24.
	LI Huaqiang, LI Xuxiang, KAN Lifeng. Market operating model and key technologies of integrated energy services under the background of energy Internet[J]. Advanced Engineering Sciences, 2020, 52(4):13-24.
[2]	赵晋泉, 夏雪, 刘子文, 等. 电力用户用电特征选择与行为画像[J]. 电网技术, 2020, 44(9):3488-3496.
	ZHAO Jinquan, XIA Xue, LIU Ziwen, et al. User electricity consumption feature selection and behavioral portrait[J]. Power System Technology, 2020, 44(9):3488-3496.
[3]	万灿, 贾妍博, 李彪, 等. 城镇能源互联网能源交易模式和用户响应研究现状与展望[J]. 电力系统自动化, 2019, 43(14):29-40.
	WAN Can, JIA Yanbo, LI Biao, et al. Research status and prospect of energy trading mode and user demand response in urban energy Internet[J]. Automation of Electric Power Systems, 2019, 43(14):29-40.
[4]	陆晓, 徐春雷, 冷钊莹, 等. 基于数据驱动方法的疫情阶段电力用户负荷特性画像模型[J]. 电力建设, 2021, 42(2):93-106.
	LU Xiao, XU Chunlei, LENG Zhaoying, et al. Load characteristic portrait model of power users in epidemic stage applying data-driven method[J]. Electric Power Construction, 2021, 42(2):93-106.
[5]	PANAPAKIDIS I P, ALEXIADIS M C, PAPAGIANNIS G K. Electricity customer characterization based on different representative load curves[C]// 2012 9th International Conference on the European Energy Market. Florence, Italy: IEEE, 2012: 1-8.
[6]	METS K, DEPUYDT F, DEVELDER C. Two-stage load pattern clustering using fast wavelet transformation[J]. IEEE Transactions on Smart Grid, 2016, 7(5):2250-2259. doi: 10.1109/TSG.2015.2446935 URL
[7]	苏适, 李康平, 严玉廷, 等. 基于密度空间聚类和引力搜索算法的居民负荷用电模式分类模型[J]. 电力自动化设备, 2018, 38(1):129-136.
	SU Shi, LI Kangping, YAN Yuting, et al. Classification model of residential power consumption mode based on DBSCAN and gravitational search algorithm[J]. Electric Power Automation Equipment, 2018, 38(1):129-136.
[8]	唐子卓, 刘洋, 许立雄, 等. 基于负荷数据频域特征和LSTM网络的类别不平衡负荷典型用电模式提取方法[J]. 电力建设, 2020, 41(8):17-24.
	TANG Zizhuo, LIU Yang, XU Lixiong, et al. Imbalanced-load pattern extraction method based on frequency domain characteristics of load data and LSTM network[J]. Electric Power Construction, 2020, 41(8):17-24.
[9]	林顺富, 顾乡, 汤继开, 等. 基于稀疏自动编码器神经网络的负荷曲线分类方法[J]. 电网技术, 2020, 44(9):3508-3515.
	LIN Shunfu, GU Xiang, TANG Jikai, et al. Power load profile classification method based on neural network of sparse automatic encoder[J]. Power System Technology, 2020, 44(9):3508-3515.
[10]	刘洋, 刘洋, 许立雄, 等. 计及数据类别不平衡的海量用户负荷典型特征高性能提取方法[J]. 中国电机工程学报, 2019, 39(14):4093-4104.
	LIU Yang, LIU Yang, XU Lixiong, et al. A high performance extraction method for massive user load typical characteristics considering data class imbalance[J]. Proceedings of the CSEE, 2019, 39(14):4093-4104.
[11]	TUMER K, GHOSH J. Error correlation and error reduction in ensemble classifiers[J]. Connection Science, 1996, 8(3/4):385-404. doi: 10.1080/095400996116839 URL
[12]	CHEN L, LU L, FENG K R, et al. Multiple classifier integration for the prediction of protein structural classes[J]. Journal of Computational Chemistry, 2009, 30(14):2248-2254.
[13]	张春霞, 张讲社. 选择性集成学习算法综述[J]. 计算机学报, 2011, 34(8):1399-1410.
	ZHANG Chunxia, ZHANG Jiangshe. A survey of selective ensemble learning algorithms[J]. Chinese Journal of Computers, 2011, 34(8):1399-1410.
[14]	王德文, 杨凯华. 基于生成式对抗网络的窃电检测数据生成方法[J]. 电网技术, 2020, 44(2):775-782.
	WANG Dewen, YANG Kaihua. A data generation method for electricity theft detection using generative adversarial network[J]. Power System Technology, 2020, 44(2):775-782.
[15]	李想, 王鹏, 刘洋, 等. 考虑类别不平衡的海量负荷用电模式辨识方法[J]. 中国电机工程学报, 2020, 40(1):128-137, 380.
	LI Xiang, WANG Peng, LIU Yang, et al. Massive load pattern identification method considering class imbalance[J]. Proceedings of the CSEE, 2020, 40(1):128-137,380.
[16]	曹鹏. 不均衡数据分类方法的研究[D]. 沈阳: 东北大学, 2014.
	CAO Peng. The research of imbalanced data classification[D]. Shenyang: Northeastern University, 2014.
[17]	BUNKHUMPORNPAT C, SINAPIROMSARAN K, LURSINSAP C. DBSMOTE: density-based synthetic minority over-sampling TEchnique[J]. Applied Intelligence, 2012, 36(3):664-684. doi: 10.1007/s10489-011-0287-y URL
[18]	刘磊, 王燕燕, 申春, 等. Bellman-Ford算法性能可移植的GPU并行优化[J]. 吉林大学学报(工学版), 2015, 45(5):1559-1564.
	LIU Lei, WANG Yanyan, SHEN Chun, et al. Performance portable GPU parallel optimization technique on Bellman-Ford algorithm[J]. Journal of Jilin University (Engineering and Technology Edition), 2015, 45(5):1559-1564.
[19]	LEE H, KIM J, KIM S. Gaussian-based SMOTE algorithm for solving skewed class distributions[J]. International Journal of Fuzzy Logic and Intelligent Systems, 2017, 17(4):229-234. doi: 10.5391/IJFIS.2017.17.4.229 URL
[20]	SAK H, SENIOR A, BEAUFAYS F. Long short-term memory recurrent neural network architectures for large scale acoustic modeling[C]// Proceedings of the Annual Conference of the International Speech Communication Association Interspeech 2014.ISCA, 2014: 338-342.
[21]	TAKAGI S, YOSHIDA Y, OKADA M. Impact of layer normalization on single-layer perceptron-statistical mechanical analysis[J]. Journal of the Physical Society of Japan, 2019, 88(7):074003. doi: 10.7566/JPSJ.88.074003 URL
[22]	谢谦, 董立红, 厍向阳. 基于Attention-GRU的短期电价预测[J]. 电力系统保护与控制, 2020, 48(23):154-160.
	XIE Qian, DONG Lihong, SHE Xiangyang. Short-term electricity price forecasting based on Attention-GRU[J]. Power System Protection and Control, 2020, 48(23):154-160.
[23]	BORAH P, GUPTA D. Functional iterative approaches for solving support vector classification problems based on generalized Huber loss[J]. Neural Computing and Applications, 2020, 32(13):9245-9265. doi: 10.1007/s00521-019-04436-x URL