基于最小二乘原理的电力变压器绝缘油色谱在线监测系统校准方法

A Calibration Method of Power Transformer Insulating Oil Chromatography Online Monitoring System Based on Least Square Principle

将半导体传感器应用于电力变压器绝缘油色谱在线监测系统时,传统的直线拟合校准方法会出现测量误差超过限值、拟合气体浓度为负值等问题。基于此,本文采用基于最小二乘原理的直线、幂函数以及一元二次函数拟合方法,对半导体传感器检测的H2、CH4、C2H6、C2H4、C2H2、CO、CO2等7种绝缘油中溶解气体的浓度数据进行拟合。通过理论分析和实验验证,证实相比直线和幂函数拟合方法,基于最小二乘原理的一元二次函数拟合结果的相关性更好,拟合误差更低,有望用于改进绝缘油色谱在线监测系统的现场校准方法。

With the application of semiconductor sensors in the online monitoring system of power transformer insulating oil chromatogram,the traditional linear fitting calibration method appears some problems,such as the measurement error exceeding the limit and the fitting gas concentration being negative and so on.In this paper,the linear,power function and one-variable quadratic function fitting method based on the least-square principle is used to fit the concentration data of seven gases(H2,CH4,C2H6,C2H4,C2H2,CO,CO2)dissolved in insulating oil detected by semiconductor sensor.Combined with theoretical analysis and experimental verification,it is proved that compared with the linear and power function fitting methods,the fitting results of the one-variable quadratic function based on the least square principle have better correlation and lower fitting error.It is expected to be used to improve the on-site calibration method of the insulating oil chromatography online monitoring system.