A Large-Pulsating-Current Generation and its Application to Identifying Parameters in Jiles-Atherton Model for Current Transformer

Abstract

The purpose of this paper is to present a method for identifying Jiles-Atherton’s parameters of current transformer (CT) using differential evolution algorithm and a pulsating current source. The pulse currents are produced from a simple inductor circuit under switched transient. As peak of inrush current can be as high as twenty times its steady-state value and lasts just one to two cycles, such approach has the advantage that it is simple yet it can produce very large current with limited investment to characterize commercial magnetic cored CTs and optical CTs with large current ratings. To illustrate the method of identifying parameters in Jiles-Atherton model using pulse current source, a current transformer with relatively low current rating was used in the experiment. Besides parameters in the Jiles-Atherton model, effective mean length, effective mean cross section and leakage inductance of the CT are also parameters to be identified. Two objective functions were used in the parameter identification, one of which uses the difference of the calculated and measured output currents and the other of which uses the difference of the calculated and measured input currents. With the identified parameters, measured input current is used to predict output current, which is found in good agreement with its measured ones; vice versa, the measured output current is used to predict the input current, which is also found in good agreement with its measured ones.  The proposed method is useful in compensation of current transformer, especially under severe fault condition, in which measured current at secondary side is significantly different from its primary side downscaled values. For protection system to operate properly, a proper compensation can be implemented by adapting the proposed method in this paper.