A method to analyze the excitation characteristics of current transformer based on Moore-Penrose inverse algorithm
Year: 2011
Type: article
Abstract: Because of the voltage-ampere curve offered by manufacturer,mostly covering the low excitation current area ranging from 5 to 200 mA,it is difficult to assess the working performance of current transformer when its primary current reaches high times of the rated value.This paper takes a simple exponent function to approach the original excitation curve and recognizes the parameters of the exponent function by pseudo-inverse algorithm.But,without data pretreatment,the parameters recognized by Moore-Penrose method will produce a big fitting error.Through regularizing the sampling interval of argument and filling few of dependent variables by linear or spline interpolation to pretreat the raw data,the influence of the nonlinear factor in recognition process is depressed and the fitting accuracy is improved.Then using the starting saturation point of excitation voltage as the reference,the dimension of parameter space which needs to be recognized can be reduced by normalized operation,and the calculation is correspondingly simplified.On the basis of fitting function gotten by recognitian,it is easy to draw the limitation curve of secondary load of current transformer under the 10% error desire.The curve is helpful to judge whether the given current transformer's secondary load matches its excitation characteristics.
Author: Yang Zhi-qiang
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2369332145