Analysis And Optimization Of Metering Accuracy Consistency Of Smart Power Meter Considering Temperature Effect

Intelligent power meter is a meter that measures energy.With the development of the global smart grid and the increasing power demand of grid users,the meter is no longer present as a single billing meter,but is moving towards intelligent,systematic,modular and diversified terminal equipment.The metering accuracy of watt-hour meters has received extensive attention from all aspects of life.The latitude range of China is wide,the temperature difference between the south and the north is large,and the work environment of the power meter is complex.The consistency of the measurement accuracy within the full working temperature range has not been paid attention to.Therefore,this article will use a single-phase smart power meter as an example,and conduct in-depth research on the simulation model of the power meter,temperature consistency assessment method,tolerance optimization method,and temperature compensation method to optimize its measurement accuracy over the full temperature range consistency.First,we set up a power meter thermoelectric synthesis simulation platform.Through the Saber simulation of the main heating components of the power meter,the power loss is obtained.The finite element model of the smart power meter was built in ANSYS Icepak,and the power loss and material properties obtained by electric simulation were input to obtain the temperature field distribution of the power meter under different ambient temperatures.According to the power meter schematic,the measurement module simulation model is built in MATLAB Simulink,and the temperature simulation result and the component temperature model are integrated into the Simulink model to realize the simulation of the meter temperature and metering function under different ambient temperatures.Secondly,a comprehensive consideration of temperature noise and manufacturing noise evaluates the consistency of metering accuracy.For its key measurement components measuring chip,voltage divider resistance,measuring temperature characteristics and build a temperature model to obtain the actual component parameters at different temperatures.Combining the tolerance values of components and components,the consistent distribution of metering accuracy can be obtained through Monte Carlo simulation.Then,temperature is used as a noise factor,and component tolerances are used as horizontal values to construct an orthogonal test table for sample sampling,and Monte Carlo simulations are used to obtain the signal-to-noise ratio of temperature for consistency evaluation of each tolerance combination.The component cost is taken into consideration and the performance-cost evaluation is obtained.The Kriging model was used to construct the approximate model of performance-cost evaluation and component tolerance.The branch-and-bound algorithm was used to solve the optimal solution of the model to obtain the best tolerance combination.Finally,the temperature drift of component parameters caused by temperature is compensated in real time in the MCU through temperature compensation.The BP neural network,RBF neural network,response surface method and Kriging method were used to set up the mapping model between the temperature of MCU and the temperature field of the metering circuit respectively.The best fitting method was selected to achieve the key point temperature of the metering circuit through the MCU temperature.Function,combined with the power meter metering principle and theoretical calculation,to obtain the temperature meter temperature compensation coefficient at different temperatures.The optimization effect is verified through simulation and actual production.