Based On Hadoop PMU And SCADA Data Hybrid State Estimation Algorithm

Nowdays,the intelligent and digitization becomes the new characters of the power network instead of informatization and intelligent.How to process,storage and use the data effectively,which is generated by electrical power system,is a problem which attracts the wide spread attention.The BigData Technology could be one of the efficient tools to handle the difficulties of processing massive data of the power system.The research objective of this paper is the data fusion of SCADA and PMU by using the BigData in electrical power system.The method of data processing of Hadoop is used in the state estimation of power system.By using the MapReduce which is the programming model of Hadoop,the algorithms of data preprocessing,bad data identification and data fusion are programmed,so that the Big Data Technology is used in the state estimation of power system.In the aspect of data preprocess,the data cleaning,data delet-repetition and data complement are implemented by MapReduce algorithm,meanwhile the algorithm also can detect and identify the bad data based on the theory of residual search method.This paper presents the two steps to identify bad data,which are detection phase and identification phase.The detection phase use real-time data to detect the abnormal data,and the identification phase use history data to identify the bad data and fault data,this method could guarantee the effectiveness of the bad data processing and avoid the phenomenon of residual contamination and residual submerge.In state estimation,in view of state estimation of data fusion of SCADA and PMU,this paper introduces a method of data fusion,by analyzing data characteristics of SCADA and PMU.Firstly,the least squares estimation algorithm is used to obtain the static state estimation data of SCADA,and then use the MapReduce Kalman Filter(MRKF)algorithm of data fusion to merge the data of SCADA and PMU,thus the precision of state estimation is improved.By using the pseudo distributed computing environment of Hadoop for experiments,the feasibility and effectiveness of the algorithms of data preprocessing,data fusion are proved,and three algorithms are integared into one algorithm,the experiment cluster of Hadoop is built to experiment the computational efficiency and computational efficiency of this algorithm.The experimental results show that the data preprocessing and MRKF algorithms,which are programmed by MapReduce,are feasible,and the calculation speed is faster at the same time,the accuracy is also guaranteed,the result prove that using big data technology in the state estimation of power system is feasible.