基于用电负荷的缺失数据插补方法研究

路昂

doi:10.16513/j.2096-2185.DE.2004018

PDF(1256 KB)

分布式能源 ›› 2020, Vol. 5 ›› Issue (4) : 74-80. DOI: 10.16513/j.2096-2185.DE.2004018

学术研究

基于用电负荷的缺失数据插补方法研究

路昂

作者信息 +

Research on Missing Data Interpolation Method Based on Electricity Load

Ang LU

Author information +

文章历史 +

摘要

电力市场改革在我国开展以后，准确的负荷预测对于电力市场需求分析具有重要意义，而采集、统计过程中的缺失数据直接影响着电力负荷预测等数据分析的精度，为此对基于用电负荷的缺失数据插补方法展开了研究。首先选取了典型工商业用户，生成用电负荷曲线，按照拟合曲线特征，对用户负荷曲线进行分类。然后随机生成了每个用户10%缺失率下的不完整数据集，并利用均值插补、回归插补和期望最大化(expectation maximization，EM)插补方法补全缺失数据。最后对比了插补后数据集与原始数据集的数据情况，通过计算均方误差(mean square error，MSE)值比较并分析了插补效果。仿真结果验证了插补方法对于不同类别用电负荷的适用性与可行性。

Abstract

After the reform of power market has been carried out in China, accurate load forecasting is of great significance for power market demand analysis. And the missing data in the process of collection and statistics directly affects the accuracy of data analysis such as power load forecasting. Therefore, the missing data interpolation method based on electricity load was studied in this paper. Firstly the typical industrial and commercial users were selected to generate the power load curves. According to the characteristics of the fitting curves, the user load curves were divided into two categories. Then each user's incomplete data sets at 10% missing rate were randomly generated. And the missing data was supplemented by means of mean interpolation, regression interpolation and EM interpolation. Finally, the data sets after interpolation were compared with the original data sets, and the effect of interpolation was analyzed and compared by calculating mean square error (MSE). The simulation results verified the applicability and feasibility of the interpolation methods for different types of electrical loads.

导出引用

路昂. 基于用电负荷的缺失数据插补方法研究[J]. 分布式能源. 2020, 5(4): 74-80 https://doi.org/10.16513/j.2096-2185.DE.2004018

Ang LU. Research on Missing Data Interpolation Method Based on Electricity Load[J]. Distributed Energy Resources. 2020, 5(4): 74-80 https://doi.org/10.16513/j.2096-2185.DE.2004018

中图分类号： TM743

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	陈雨果，李嘉龙，刘文涛，等. 电力市场环境下发电计划关键技术综述[J]. 广东电力，2018, 31(6): 8-13. CHEN Yuguo, LI Jialong, LIU Wentao, et al. Overview on key technologies for generation scheduling under electric market environment[J]. Guangdong Electric Power, 2018, 31 (6): 8-13. 本文引用 [1]

[2]	陈毅平，覃智君. 电力市场改革研究综述[J]. 电工电气，2019(4): 1-6, 12. CHEN Yiping, QIN Zhijun. Review of research on electricity market reform[J]. Electrotechnics Electric, 2019(4): 1-6, 12. 本文引用 [1]

[3]	白杨，李昂，夏清. 新形势下电力市场营销模式与新型电价体系[J]. 电力系统保护与控制，2016, 44(5): 10-16. BAI Yang, LI Ang, XIA Qing. Electricity business marketing modes in the new environment and new electricity pricing systems[J]. Power System Protection and Control, 2016, 44(5): 10-16. 本文引用 [1]

[4]	王建军. 智能挖掘电力负荷预测研究及应用[M]. 北京：中国水利水电出版社，2013. WANG Jianjun. Research and application of intelligent mining power load prediction[M]. Beijing: China Water & Power Press, 2013. 本文引用 [1]

[5]	张建寰，吉莹，陈立东. 深度学习在电力负荷预测中的应用[J]. 自动化仪表，2019, 40(8): 8-12, 17. ZHANG Jianhuan, JI Ying, CHEN Lidong. Application of deep learning in power load forecasting[J]. Process Automation Instrumentation, 2019, 40(8): 8-12, 17. 本文引用 [1]

[6]	马哲，舒勤. 基于ESPRIT分解算法的短期电力负荷预测[J]. 电力系统保护与控制，2015, 43(7): 90-96. MA Zhe, SHU Qin. Short term load forecasting based on ESPRIT integrated algorithm[J]. Power System Protection and Control, 2015, 43(7): 90-96. 本文引用 [1]

[7]	张栋梁，严健，李晓波，等. 基于马尔可夫链筛选组合预测模型的中长期负荷预测方法[J]. 电力系统保护与控制，2016, 44(12): 63-67. ZHANG Dongliang, YAN Jian, LI Xiaobo, et al. Mid-long term load forecasting based on Markov chain screening combination forecasting models[J]. Power System Protection and Control, 2016, 44(12): 63-67. 本文引用 [1]

[8]	杨慧霞，邓迎君，刘志斌，等. 含有历史不良数据的电力负荷预测研究[J]. 电力系统保护与控制，2017, 45(15): 62-68. YANG Huixia, DENG Yingjun, LIU Zhibin, et al. Study on electric load forecasting with historical bad data[J]. Power System Protection and Control, 2017, 45(15): 62-68. 本文引用 [1]

[9]	庞新生. 缺失数据插补处理方法的比较研究[J]. 统计与决策，2012(24): 18-22. PANG Xinsheng. Comparative research on imputation processing methods for missing data[J]. Statistics and Decision, 2012(24): 18-22. 本文引用 [1]

[10]	XIE Mingshan, HUANG Mengxing, BAI Yong, et al. A robust data interpolation based on a back propagation artificial neural network operator for incomplete acquisition in wireless sensor networks[J]. Journal of Sensors, 2018: 1-16. 本文引用 [1]

[11]	邱贻涛，吴刘仓，马婷. 缺失数据下联合均值与方差模型的参数估计[J]. 数理统计与管理，2015, 34(4): 621-627. QIU Yitao, WU Liucang, MA Ting. Parameter estimation of the joint mean and variance model under missing data[J]. Mathematical Statistics and Management, 2015, 34(4): 621-627. 本文引用 [1]

[12]	陈慧佳. 基于Random Forest的缺失数据补全策略研究[D]. 南昌：南昌大学，2016. CHEN Huijia. Research on strategy of imputing missing data based on random forest[D]. Nanchang: Nanchang University, 2016. 本文引用 [1]

[13]	杨晓倩. 缺失数据插补方法的选择研究[D]. 兰州：兰州财经大学，2016. YANG Xiaoqian. Research on interpolation method of missing data[D]. Lanzhou: Lanzhou University of Finance and Economics, 2016. 本文引用 [1]

[14]	王敏. 基于成分数据的缺失值补全方法研究[D]. 太原：山西大学，2016. WANG Min. Imputation methods of missing values for compositional data[D]. Taiyuan: Shanxi University, 2016. 本文引用 [1]