Models for capacitive effects in iron core transformers

Abstract

The paper presents an efficient modeling technique for iron-core coils and transformers based on a separation of the nonlinear inductive effects from the linear capacitive effects. The magnetic field is numerically analyzed neglecting the displacement currents. The distributed capacitive effects in the insulation between coils are concentrated in an "Extended 5 scheme," an infinite circuit which is optimally reduced to a finite one. The separately extracted models are integrated in a common discrete (lumped) system. This global model obtained by postprocessing the numerical field solution was used to study the transmission of atmospheric lightning overvoltages to the secondary windings of power transformers.

Modelling of foil‐type transformer windings for computation of terminal impedance and internal voltage propagation

“…To investigate the core influence, Z w ( jω) is computed by three different methods; (i) by taking into account zero magnetic flux penetration into the magnetic core using (8), (ii) by taking into account the magnetic core effects applying formula (9) and using constant parameter values ρ = ρ′ = 4.8 × 10 −7 Ωm and μ z = 4.56 × 10 −4 H/m and (iii) by taking into account the magnetic core effects by applying formulas (9) and frequency dependent parameters using (19) and (20) and the values for ρ =10 −1 Ωm, a =10 −10 Ωms/rad, k = 0.628 (Ωs 0.3 rad 0.7 )/m and n = −0.7. The results are shown in Fig.…”

“…The model with frequency dependent parameters takes into account the effect of the magnetic core and the frequency dependency of the parameters ρ, ρ′ and μ z of the magnetic material. It is capable to produce accurate results for frequencies approximately below 25 kHz till approximately 5 kHz which is the boundary for the assumptions made for ρ′ and μ z through (19) and (20), respectively. To improve the computations using the frequency dependent parameter model, the method in (9) in combination with refine representation of the parameters ρ, ρ′ and μ z can be used.…”

“…Either for the distributed parameter or for the lumped parameter models, parameter determination is very important because it heavily affects model accuracy. In [19][20][21][22][23][24][25], accurate computation of parameters by finite elements methods has been reported. Apart from this, there are efficient computational methods that can be used for the determination of the inductance and capacitance matrix as well as accurate representation of losses.…”