The Improvement of Calibration Method of Power Transformer Oil Chromatography Online Monitoring System

Abstract

Abstract When the semiconductor sensors are applied to the power transformer oil chromatography online monitoring system, the error of the traditional linear fitting calibration method exceeds the limit value. Therefore, this paper carries out the research of improving the calibration method of oil chromatography online monitoring system, by calibrating the online monitoring values of seven dissolved gases in reference oil samples using linear function, quadratic function and power function models, respectively. Besides, by evaluating the fitting effect, it is confirmed that the quadratic function fitting result has better correlation and lower fitting error, which is expected to be used to improve the field calibration method of the oil chromatography online monitoring system.