On Typical Flaws of the Transformer Models for Inrush Current Evaluation

For many decades, the main, if not the only, equivalent circuit of a single-phase transformer was its T-model. To avoid ambiguity and take into account the topology of the core and windings, we will use the terms inner and outer windings, instead of the primary and secondary windings. So, inductanc...

Full description

Saved in:
Bibliographic Details
Main Authors: Serhii Zirka, Yurii Moroz
Format: Article
Language:English
Published: Oles Honchar Dnipro National University 2023-06-01
Series:Challenges and Issues of Modern Science
Online Access:https://cims.fti.dp.ua/j/article/view/59
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:For many decades, the main, if not the only, equivalent circuit of a single-phase transformer was its T-model. To avoid ambiguity and take into account the topology of the core and windings, we will use the terms inner and outer windings, instead of the primary and secondary windings. So, inductances LS1 and L′S2 are the so-called "leakage inductances" of the inner and outer windings, respectively, r1 and r′2 are their resistances. In order to reproduce hysteretic (quasi-static) properties of the core material and to account for the dynamic core losses and saturation, the magnetizing branch is represented by an ATP-inductor DHM, which implements, starting with 2019 version, a Dynamic Hysteresis Model (DHM). We have attracted attention to typical mistakes made in modeling transformer inrush currents. It is explained that the main mistake is the use of the convenient T-model and their three-phase derivatives. Their disadvantages are rooted in using separate leakage models of primary and secondary circuits, as well as in inability to reproduce different magnetization levels of the legs and yoke. We show the benefits of the π-model, but at the same time, caution against its oversimplification. Although we use an advanced hysteresis model to describe processes in the legs and yokes, it is shown that accounting hysteresis and core losses is completely optional in modeling inrush currents. These model properties have no visible impact neither on the current peaks nor their decay in time.
ISSN:3083-5704