Analyzing And Application Based On Big Data Of Smart Energy Meter

In recent years,the State Grid of China has gradually accelerated the construction of smart grids.As a core part,State Grid of China has a large number of latitudes for the accumulation of smart meter data.In recent years,data analysis has an important influence in various fields.The application of data analysis has provided new thinking and management system models for improving the asset management technology and level of energy meters,and status assessment.Our work is based on smart meter data.In this paper,we use machine learning and other data analysis methods to model the smart meter supplier,meter fault and life prediction,and predict meter error based on the tree structure.The energy meter supplier modeling use EM algorithm-based iterative method to separate the energy meter suppliers and installation areas of the impact of the failure.Smart meter fault and life prediction based on the modeling of manufacturers and regions,using a variety of machine learning method to get meter fault and life analysis model.The remote error detect of the smart meter calculates the error of all electric energy meters based on the electricity records.Based on the original method,the calculating method is extended and improved,so that the method can be used in the actual area.The above work has achieved fine results in practical use.In the case of energy meter suppliers and regional failure modelling and power meter failure prediction and lifetime prediction,the results we obtained are match with the actual situation.The remote error calculation method of the electric energy meter has been verified by the laboratory,small-scale on-site verification,large-scale verification,and all obtained correct results.