Data cleaning method for distribution transformer

被引：0
|
作者：
Liu Y. [1 ]
Luan W. [2 ]
Xu Y. [1 ]
Wang P. [2 ]
Guo S. [2 ]
机构：
[1] School of Electrical Engineering, North China Electric Power University, Baoding, 071000, Hebei Province
[2] China and Electronic Electric Power Research Institute, Haidian District, Beijing
来源：
| 1600年 / Power System Technology Press卷 / 41期
关键词：
Big data; Data cleaning; Distribution transformer; Spark technology;
D O I：
10.13335/j.1000-3673.pst.2016.1513
中图分类号：
学科分类号：
摘要：
Due to wide distribution, complex grid structure, poor operating environment and other objective conditions, communication reliability in power distribution/utilization system is difficult to guarantee. In power distribution/ utilization system, there are many abnormal and missing data in actual operation in distribution transformer. Actual data of operational distribution transformer are used as study sample in this paper. Reasons of abnormal and missing data are analyzed. Then algorithm of small sample data is used. A method based on kernel smoothing technology is proposed to solve abnormal data, and a method based on Pearson correlation coefficients and regression model is proposed to recover missing data. In order to verify efficiency of the proposed method, Spark parallel computing architecture is built with six servers. Abnormal data identification and missing data recovery is carried out on TB level. Related results prove practical value of the proposed method. © 2017, Power System Technology Press. All right reserved.
引用
收藏
页码：1008 / 1014
页数：6
相关论文
共 17 条
  • [1] Zhang D., Miao X., Liu L., Et al., Research on development strategy for smart grid big data, Proceedings of the CSEE, 35, 1, pp. 2-12, (2015)
  • [2] Zhao T., Zhang Y., Zhang D., Application technology of big data in smart distribution grid and its prospect analysis, Power System Technology, 38, 12, pp. 3305-3312, (2014)
  • [3] Liu K., Sheng W., Zhang D., Et al., Big data application requirements and scenario analysis in smart distribution network, Proceedings of the CSEE, 35, 2, pp. 287-293, (2010)
  • [4] Wang Y., Le C., A data pretreatment technique about power system load modeling, Power System Technology, 31, pp. 292-294, (2007)
  • [5] Zhang P., Wu X., He J., Review on big data technology applied in active distribution network, Power Construction, 36, 1, pp. 52-59, (2015)
  • [6] Yan Y., Sheng G., Chen Y., Et al., An method for anomaly detection of state information of power equipment based on big data analysis, Proceedings of the CSEE, 35, 1, pp. 52-59, (2015)
  • [7] Luan W., Wang B., Zhou N., Et al., Distribution network based on metering data, Power System Technology, 39, 11, pp. 3141-3146, (2015)
  • [8] Zhao L., Luan W., Wang Q., Accuratea line loss analysis of LV distribution network using AMI data, Power System Technology, 39, 11, pp. 3189-3194, (2015)
  • [9] Diao Y., Sheng W., Liu K., Et al., Research on online cleaning and repair methods of large-scale distribution network load data, Power System Technology, 39, 11, pp. 3134-3140, (2015)
  • [10] Liu P., Lei L., Zhang X., Missing data processing method of comparative study, Journal of Computer Science, 9, pp. 155-156, (2004)
12