A comparative study of J–A and Preisach improved models for core loss calculation under DC bias

Yang Li,Zhichao Luo
DOI: https://doi.org/10.1038/s41598-024-55155-w
IF: 4.6
2024-04-25
Scientific Reports
Abstract:Power transformer is widely used in the power system with the rapid development of the power converters connected to the grid. When a transformer operates under DC bias conditions, its iron core loss increases significantly, causing local overheating and threatening the proper operation of the transformer. However, there are persistent difficulties in accurately assessing the core loss when the induction waveform is influenced by a DC bias. This paper first proposes improvements to the J–A and Preisach models to evaluate the core loss of the iron core under DC bias. Additionally, we incorporate the hysteresis models into the finite element method (FEM) by modifying the fundamental constitutive equations in the FEM model in order to perform a precise core loss/distribution calculation. To verify the accuracy of prediction, a transformer prototype with a laminated core is developed. The improved J–A-FEM and Preisach-FEM models were directly compared in terms of calculation accuracy, numerical implementation, and computational burden.
multidisciplinary sciences
What problem does this paper attempt to address?
This paper attempts to solve the problem of accuracy in evaluating the core loss of transformers under DC - bias conditions. Specifically: 1. **Background problems**: - With the increasing demand for power converters to be connected to the power grid, transformers are more and more widely used in power systems. - When the transformer operates under DC - bias conditions, the core loss increases significantly, resulting in local overheating and threatening the normal operation of the transformer. - However, when the induction waveform is affected by DC - bias, there are continuous difficulties in accurately evaluating the core loss. 2. **Limitations of existing methods**: - Traditionally, the core loss is predicted by empirical formulas based on the Steinmetz equation, and the parameters of these formulas are determined by experimental measurements under sinusoidal excitation. - Although the existing improved empirical formulas provide better accuracy in core loss estimation, they still rely on experimental measurements. - The influence of DC - bias magnetization on the core loss is significant and complex, and it is difficult to obtain the hysteresis loops and core losses under all different DC - bias conditions to fit the Steinmetz equation. Therefore, the Steinmetz method is neither economical nor time - saving and cannot be used as a general method for accurately calculating the core loss. 3. **Research objectives**: - This paper first proposes improvements to the Jiles - Atherton (J - A) model and the Preisach model to evaluate the core loss of the core under DC - bias conditions. - In addition, by modifying the basic constitutive equations in the finite element method (FEM) model, the hysteresis model is incorporated into the FEM to perform accurate core loss distribution calculations. - To verify the accuracy of the prediction, a prototype of a laminated - core transformer was developed, and the performance of the improved J - A - FEM and Preisach - FEM models in terms of calculation accuracy, numerical implementation, and computational burden was directly compared. 4. **Main contributions**: - Improved J - A and Preisach models are proposed to more accurately describe the core loss under DC - bias conditions. - The improved hysteresis model is combined with the finite element method to improve the accuracy of core loss calculation. - The accuracy and applicability of the models are verified by experiments, especially the performance under DC - bias conditions. In conclusion, this paper aims to improve the accuracy of evaluating the core loss of transformers under DC - bias conditions by improving the existing hysteresis models and combining them with the finite element method.