Optimized high-frequency white-box transformer model for implementation in ATP-EMTP

Abstract

Although several high frequency white-box transformer models have been proposed in the literature, the methodology for their implementation in some programs based on EMTP has not been addressed in detail. This is true in particular for ATP, for which the representation of large transformer models is particularly problematic. This work aims to enable the use of ATP to model detailed transformer equivalent circuits, which would otherwise be seriously questioned. To establish these circuit models, the following must be taken into account: model reduction, magnetic circuit decoupling, avoid numerical instability, limit and optimize the use of resistors, generate ATP versions capable of processing large models. To this end, three generally valid methodologies are presented to reduce the model without compromising the accuracy of the calculation results and, finally, additional specific methodologies are implemented to reduce and adapt the models in the ATP environment. The application of these methodologies results in six possible variants of reduced models, which were successfully validated through a case study used by CIGRE JWG A2/C4.52 to verify the accuracy of transformer models.