计算机科学 ›› 2021, Vol. 48 ›› Issue (2): 128-133.doi: 10.11896/jsjkx.191200152

• 数据库&大数据&数据科学 • 上一篇    下一篇

基于稀疏表示的电力负荷数据补全

李培冠1, 於志勇1,2, 黄昉菀1,2   

  1. 1 福州大学数学与计算机科学学院 福州350116
    2 福州大学福建省网络计算与智能信息处理重点实验室 福州350116
  • 收稿日期:2019-12-25 修回日期:2020-04-16 出版日期:2021-02-15 发布日期:2021-02-04
  • 通讯作者: 黄昉菀(hfw@fzu.edu.cn)
  • 作者简介:lipeiguan163@163.com
  • 基金资助:
    国家自然科学基金(61772136);福建省中青年教师教育科研项目(JT180045)

Power Load Data Completion Based on Sparse Representation

LI Pei-guan1, YU Zhi-yong1,2, HUANG Fang-wan1,2   

  1. 1 College of Mathematics and Computer Science,Fuzhou University,Fuzhou 350116,China
    2 Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing,Fuzhou University,Fuzhou 350116,China
  • Received:2019-12-25 Revised:2020-04-16 Online:2021-02-15 Published:2021-02-04
  • About author:LI Pei-guan,born in 1996,B.S.His main research interests include machine learning and so on.
    HUANG Fang-wan,born in 1980,M.S.,senior lecturer,is a member of China Computer Federation.Her main research interests include computational intelligence,machine learning and big data analysis.
  • Supported by:
    The National Natural Science Foundation of China(61772136)and Research Project for Young and Middle-aged Teachers of Fujian Province (JT180045).

摘要: 数据缺失在电力负荷数据采集过程中经常发生,对提高算法的预测精确度带来了不利影响。现有的缺失数据补全算法只适用于缺失数据量较少的情况,而对于缺失数据较多的情况表现不佳。面对严重数据缺失的挑战,文中提出了一种基于稀疏表示的电力负荷缺失数据补全方法。首先以数据随机缺失为前提,将训练数据中假定缺失后的数据与完整的训练数据上下拼接构成训练矩阵;其次,利用离散余弦变换(Discrete Cosine Transform,DCT)生成一个过完备字典,并根据训练矩阵对其进行学习,旨在通过调优得到一个合适的字典,能对训练矩阵中的样本进行最好的稀疏表示。最后,在测试阶段,先利用学习后字典的上半部分获得测试集缺失数据的稀疏表示,然后利用稀疏表示和学习后字典的下半部分重构出无缺失的完整数据。实验结果表明,使用该方法对电力负荷数据缺失值进行补全,可以获得比传统插值方法、基于相关性的KNN算法、时空压缩感知估计算法以及时序压缩感知预测算法更高的精度。即使数据缺失率高达95%,该方法依然可以有效地补全缺失数据。

关键词: 电力负荷, 缺失数据, 数据补全, 稀疏表示

Abstract: Data loss often occurs in the process of power load data collection,which adversely affects the accuracy of algorithm prediction.The existing missing data completion algorithm is only suitable for the case with less missing data,but performs poorly for the case with more missing data.Faced with the challenge of severe data loss,a method for power load missing data completion based on sparse representation is proposed.First of all,we assume that the data is randomly missing,and stitch the assumed missing data in the training data and the complete training data to form a training matrix.Secondly,an over-complete dictionary is generated by discrete cosine transform (DCT),and is learned according to the training matrix,aims to obtain a suitable dictionary for the best sparse representations of the samples in the training matrix.Finally,in the test phase,the upper part of the learned dictionary is used to obtain sparse representations of the missing data in the test set,and then the sparse representations and the lower part of the learned dictionary are used to reconstruct the complete data without missing.Experimental results show that using this method to complete missing values of power load data can achieve higher accuracy than traditional interpolation me-thods,correlation-based KNN algorithm,spatiotemporal compressed sensing estimation algorithm and time-series compressed sen-sing prediction algorithm.Even if the data miss rate is as high as 95%,this method can still effectively complete the missing data.

Key words: Data completion, Missing data, Power load, Sparse representation

中图分类号: 

  • TP181
[1] FAN W Q,ZHANG W,LI Y G,et al.Ultra short-term loadforecasting for micro-grid based on improved human comfort index[J].Guangdong Electric Power,2017,30(4):137-142.
[2] LI H.Residual error GM(1,1) model improved by Markovmethod for long-term and medium-term load forecasting[J].Shaanxi Electric Power,2017,45(5):75-77.
[3] YANG H X,DENG Y J,LIU Z B,et al.Study on electric load forecasting with historical bad data[J].Power System Protection and Control,2017,45(15):62-68.
[4] CHEN Z H,ZHANG Y,WU Z G.Application of RBF neural network in medium and long-term load forecasting[J].Procee-dings of the CSU-EPSA,2006,18(1):15-19.
[5] LAKSHMINARAYAN K.Imputation of missing data in industrial databases[J].Applied Intelligence,1999,11:259-275.
[6] WU S F,CHANG C Y,LEE S J.Time series forecasting with missing values[C]//2015 1st International Conference on Industrial Networks and Intelligent Systems (INISCom).2015:151-156.
[7] DING Q,LI S.Research on Resampling Application of Intelli-gent Substation and Error Analysis of Linear Interpolation Method[J].Power System Protection and Control,2015,43(23):132-136.
[8] ZHU Q W,YE L,ZHAO Y N,et al.Research on Identification and Reconstruction Method of Wind Farm Output Power Abnormal Data[J].Power System Protection and Control,2015,43(3):38-45.
[9] RUAN Q Z,CHEN J B,ZHU G,et al.Instantaneous test data analysis of low voltage electrical equipment based on cubic spline interpolation[J].Low Voltage Electrical Appliance,2012(10):27-31.
[10] TAO T Y,WANG H.Simplified wind field simulation based on Hermite interpolation[J].Engineering Mechanics,2017,34(3):182-188.
[11] GERHARD T,SHAHLA R.Improved methods for the imputation of missing data by nearest neighbor methods[J].Computational Statistics and Data Analysis,2015,90:84-99.
[12] NEWSHAM G R,BIRT B J.Building-level occupancy data to improve arima-based electricity use forecasts[C]//Proceedings of the 2nd ACM Workshop on Embedded Sensing Systems for Energy-Efficiency in Building,ACM.New York,USA,2010:13-18.
[13] SHI W,ZHU Y,ZHANG J,et al.Improving power grid monitoring data quality:An efficient machine learning framework for missing data prediction[C]//2015 IEEE 17th International Conference on High Performance Computing and Communications.IEEE,2015:417-422.
[14] KONG L,XIA M,LIU X Y,et al.Data loss and reconstruction in sensor networks[C]//INFOCOM.2013:1654-1662.
[15] ZHU Y,LI Z,ZHU H,et al.A Compressive Sensing Approach to Urban Traffic Estimation with Probe Vehicles[J].IEEE Transactions on Mobile Computing,2013,12(11):2289-2302.
[16] SONG X X,GUO Y,LI N,et al.Missing Data Prediction Based on Compressive Sensing in Time Series[J].Computer Science,2019,46(6):35-40.
[17] ZHANG Y,ROUGHAN M,WILINGER W,et al.Spatio-tem-poral compressive sensing and internet traffic matrices[J].ACM SIGCOMM Computer Communication Review,2009,39(4):267.
[18] MEI J L,YOHANN D C,YANNIG G,et al.Nonnegative matrix factorization with side information for time series recovery and prediction[J].IEEE Transactions on Knowledge and Data Engineering,2018:1.
[19] WANG Z H,HORNG G J,HSU T H,et al.Heart sound signal recovery based on time series signal prediction using a recurrent neural network in the long short-term memory model[J].The Journal of Supercomputing,2019(1):1-18.
[20] STRAUMAN A S,BIANCHI F M,MIKALSEN K Ø.Classification of postoperative surgical site infections from blood measurements with missing data using recurrent neural networks[C]//International Conference on Biomedical & Health Informatics.Las Vegas,USA,2018:307-310.
[21] ZHANG Z,XU Y,YANG J,et al.A survey of sparse representation:algorithms and applications[J].Access IEEE,2015,3:490-530.
[22] AHARON M,ELAD M,BRUCKSTEINA.K-SVD:an algo-rithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322.
[1] 庄世杰, 於志勇, 郭文忠, 黄昉菀.
基于Zoneout的跨尺度循环神经网络及其在短期电力负荷预测中的应用
Short Term Load Forecasting via Zoneout-based Multi-time Scale Recurrent Neural Network
计算机科学, 2020, 47(9): 105-109. https://doi.org/10.11896/jsjkx.190800030
[2] 田旭, 常侃, 黄升, 覃团发.
基于残差字典及协作表达的单图像超分辨率算法
Single Image Super-resolution Algorithm Using Residual Dictionary and Collaborative Representation
计算机科学, 2020, 47(9): 135-141. https://doi.org/10.11896/jsjkx.190600146
[3] 程中建, 周双娥, 李康.
基于多尺度自适应权重的稀疏表示目标跟踪算法
Sparse Representation Target Tracking Algorithm Based on Multi-scale Adaptive Weight
计算机科学, 2020, 47(6A): 181-186. https://doi.org/10.11896/JsJkx.190500093
[4] 吴庆洪, 高晓东.
稀疏表示和支持向量机相融合的非理想环境人脸识别
Face Recognition in Non-ideal Environment Based on Sparse Representation and Support Vector Machine
计算机科学, 2020, 47(6): 121-125. https://doi.org/10.11896/jsjkx.190500058
[5] 李笑宇,高清维,卢一相,孙冬.
一种根据图像能量调整的图像融合方法
Image Fusion Method Based on Image Energy Adjustment
计算机科学, 2020, 47(1): 153-158. https://doi.org/10.11896/jsjkx.181202437
[6] 李桂会,李晋江,范辉.
自适应匹配追踪图像去噪算法
Image Denoising Algorithm Based on Adaptive Matching Pursuit
计算机科学, 2020, 47(1): 176-185. https://doi.org/10.11896/jsjkx.181202280
[7] 张冰, 谢从华, 刘哲.
基于显著稀疏表示和邻域信息的多聚焦图像融合
Multi-focus Image Fusion Based on Latent Sparse Representation and Neighborhood Information
计算机科学, 2019, 46(9): 254-258. https://doi.org/10.11896/j.issn.1002-137X.2019.09.038
[8] 宋晓祥,郭艳,李宁,余东平.
基于稀疏贝叶斯学习的协同进化时间序列缺失数据预测算法
Missing Data Prediction Algorithm Based on Sparse Bayesian Learning in Coevolving Time Series
计算机科学, 2019, 46(7): 217-223. https://doi.org/10.11896/j.issn.1002-137X.2019.07.033
[9] 王丽芳, 史超宇, 蔺素珍, 秦品乐, 高媛.
基于联合图像块聚类自适应字典学习的多模态医学图像融合
Multi-modal Medical Image Fusion Based on Joint Patch Clustering of Adaptive Dictionary Learning
计算机科学, 2019, 46(7): 238-245. https://doi.org/10.11896/j.issn.1002-137X.2019.07.036
[10] 张福旺, 苑会娟.
一种自适应稀疏表示和非局部自相似性的图像超分辨率重建算法
Image Super-resolution Reconstruction Algorithm with Adaptive Sparse Representationand Non-local Self-similarity
计算机科学, 2019, 46(6A): 188-191.
[11] 宋晓祥, 郭艳, 李宁, 王萌.
基于压缩感知的时间序列缺失数据预测算法
Missing Data Prediction Based on Compressive Sensing in Time Series
计算机科学, 2019, 46(6): 35-40. https://doi.org/10.11896/j.issn.1002-137X.2019.06.004
[12] 杜秀丽, 左思铭, 邱少明.
基于图像灰度熵的自适应字典学习算法
Adaptive Dictionary Learning Algorithm Based on Image Gray Entropy
计算机科学, 2019, 46(5): 266-271. https://doi.org/10.11896/j.issn.1002-137X.2019.05.041
[13] 茹锋, 徐锦, 常琪, 阚丹会.
一种用于影像遗传学关联分析的高阶统计量结构化稀疏算法
High Order Statistics Structured Sparse Algorithm for Image Genetic Association Analysis
计算机科学, 2019, 46(4): 66-72. https://doi.org/10.11896/j.issn.1002-137X.2019.04.010
[14] 毛峡, 王岚, 李建军.
一种基于RGB-D特征融合的人体行为识别框架
Human Action Recognition Framework with RGB-D Features Fusion
计算机科学, 2018, 45(8): 22-27. https://doi.org/10.11896/j.issn.1002-137X.2018.08.005
[15] 甘玲, 赵福超, 杨梦.
一种自适应组稀疏表示的图像修复方法
Self-adaptive Group Sparse Representation Method for Image Inpainting
计算机科学, 2018, 45(8): 272-276. https://doi.org/10.11896/j.issn.1002-137X.2018.08.049
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!