Managing and Analyzing Big Error Data for Intelligent Electric Energy Meter Quality Monitoring

Abstract

With the installation and application of intelligent electric energy meters, the need for the veracity of meters from the users is increased thereupon. How to monitor and predict the quality of intelligent electric energy meters based on big error data acquired from intelligent electric energy meters is non-trivial. However, managing and analyzing large volumes of intelligent electric energy meter error data is a great challenge and is consequently hindering the effective utilization of the big dataset collected. In this paper, techniques for collecting, storing and analyzing large volumes of error data is presented to reduce the uncertainty of intelligent electric energy meters quality estimation. First, the error data of intelligent electric energy meters is sampled automatically and accurately by a designed device called Self-Service Error Calibrator Device. Then a storage system based on fountain codes is designed to store the error data on the HDFS platform. At last, an improved data mining technology named HMM (Hidden Markov Model) is leveraged to predict the variation performance tendency of intelligent electric energy meters varied with power load, manufacturers. Simulation results show that our proposed system can provide efficient, accurate, and cost-effective big error data management for intelligent electric energy meters quality monitoring.