Stochastic Characteristic Analysis Of Nonlinear Load And Dynamic Test Signal Modeling

In recent years,with the improving of the smart grid,the nonlinear load in the grid are increasing.It has become an important part of the power system load.On the one hand,the nonlinear dynamic loads have a complex power fast random variation characteristics,while this can cause large power meter dynamic errors,and also lead to differences in measurement error.Therefore,the research on dynamic load variation characteristics,find out the nonlinear signal random dynamic load characteristics is very important.On the other hand,the characteristic of random variation of nonlinear load is an important basis for establishing the signal model of dynamic errors test for electric energy meter,and reasonable and effective dynamic test signal is the key to analyze the dynamic error characteristics of the electric energy meter.Therefore,it is also very important to study the dynamic test signal model of the energy meter which can reflect the random characteristics of nonlinear loads.Firstly,this paper introduces the dynamic load classification under smart grid,the latest research for nonlinear load characteristics in China and abroad,and the development of dynamic test signal modeling of the electric energy meter.Secondly,aiming at the problems of power system nonlinear dynamic load fundamental non-stationary signal stochastic characteristics analysis,this paper studies the method of transforming non-stationary random signals into weakly stationary random signals.A histogram curve fitting method of the optimal probability density function and an improved Kullback-Leibler distance detection method are used.The discriminant mechanism of the distribution characteristic of the load fundamental envelope signal is established.On this basis,this paper analyzes the random characteristics of the four types of typical nonlinear load fundamental current and power signal collected in the field.The random analysis results of load are given.This paper lays a foundation for improving the accuracy of electric energy metering and establishing a reasonable and effective dynamic test signal model of electric energy meter.Thirdly,according to the IEC61850 protocol,this paper establishes the m-sequence discrete dynamic test signal model which reflects the typical random characteristics of nonlinear loads.At the same time,it derives the electric energy measurement algorithm of Hadamand dot product non-overlapping weighted cumulative.The simulation results give the dynamic electric energy measurement errors caused by the quantization error of IEC61850 protocol,which provides an effective method for evaluating the dynamic error characterisitics of all digital electric energy meter.Finally,basing on WindowsCE6.0 operating platform,this paper completes the HE5025 PC software design and development.The m-sequence dynamic testing signals under random excitation in HE5025 deviceis is developed.This software provides a software foundation for the HE5025.