Electromagnetic-thermal analysis of 400 kV power transformer on ITER PPEN substation

Abstract

As an important electrical power distribution system of ITER (International Thermonuclear Experimental Reactor) device, PPEN (Pulsed Power Electrical Network) feeds the pulsed loads for the superconducting coils and the heating and current driving systems. The PPEN transforms 400 kV power grid to 66 kV and 22 kV distribution voltage levels by 3 main step-down transformers. In this paper, the power loss of PPEN transformer is analyzed and calculated. The 2D PPEN transformer model is built. The temperature rise of PPEN transformer is calculated by ANSYS software. To validate the feasibility and correctness of the proposed method, the temperature rise test of PPEN transformer is carried out, then the related results are compared with the test data.

Keywords

ITER PPEN temperature rise ANSYS short-circuit method

1. Introduction

ITER (International Thermonuclear Experimental Reactor) is the world’s largest tokamak fusion experimental device under construction until now [1–3]. Its transmission voltage and installed capacity are in the level of 400 kV/1.2 GVA/27 GW. As one of the most important electrical power distribution systems of ITER device, PPEN feeds the pulsed loads for the superconducting coils and the heating and current driving systems [4]. The key equipment of PPEN consists of the main step-down transformers, which are operated in pulsed mode. The pulse power is such huge that the maximum instantaneous active power is about 600 MW and the maximum reactive reactive power is about 900 Mvar. Such a complex condition is unprecedented ordeal for PPEN transformer.

The thermal performance is an essential factor when deciding the operating life and economic performance of power transformer [5,6]. Hence, the accurate calculation on temperature rise of power transformer and the preventive measures for local overheating are concerned by domestic and foreign researchers. As excessive temperature rise may cause local overheating and shorten the transformer’ lifespan [7], an accurate prediction of the location of hot-spot appeared in the transformer has become one of the most important issues in the design of transformer winding.

In this paper, based on the structural design of PPEN transformer, the temperature distribution and temperature rise of transformer is analyzed by ANSYS software. Firstly, the power loss including load loss and no-load loss is analyzed and calculated using theoretical formulas. Secondly, by the application of magnetic and thermal field indirect coupling calculation scheme, the power loss is converted into heat production rate which can be taken as a heat source to solve the temperature distribution of the transformer. Finally, by the comparison between experimental data and simulation results, the feasibility and correctness of the proposed model and method is validated.

2. PPEN transformer

The PPEN transformer is a three-phase three windings power transformer equipped with a three-phase five limbs core structure. The limited values of temperature rise are defined referring to the standard IEC60076.2-2011 power transformer-part 2: Temperature rise for liquid-immersed transformers. Basic parameters of the transformer are shown in Table 1.

Table 1
Basic parameters of PPEN transformer

Rated power (MVA) 210/175/105 (ONAN)

300/250/150 (ONAF)

Rated voltage (kV) 400 ± 8 × 1.25%/66/22.5

Vector group YNyn0d11

Frequency (Hz) 50

Insulation type oil-immersed type

Short-circuit Resistance HV-MV11%, HV-LV25%, MV-LV14%

Temperature rise 60/65/78K(top oil/winding average/hot-spot)

Type of cooling ONAN/ONAF

3. Description of the study

3.1. Electro magnetic and thermal field indirect coupling

The power loss of PPEN transformer includes load loss and no-load loss under the normal condition. The load loss is comprised of resistance loss, eddy-current loss and stray loss, which is the primary cause leading to the rise of the winding temperature. It can be described as follows: $\begin{eqnarray}\displaystyle P_{Cu}=I^{2}R+P_{e}+P_{S}, & & \displaystyle\end{eqnarray}$ (1) where I ² R is for resistance loss of winding; P _e is for eddy-current loss of winding; P _S is for stray loss of winding.

The no-load loss involves basic core loss and stray loss in the iron core. The basic core loss is the sum of magnetic hysteresis loss and eddy-current loss. Usually, according to the type of core silicon steel sheet and maximum core flux density (B _m), the unit weight of basic core loss can be found from relevant manuals in practice. The stray loss can be acquired by coefficient without calculation. So the expression of no-load loss is shown as follows: $\begin{eqnarray}\displaystyle P_{Fe}=K_{p0}\times G_{Fe}\times p_{Fe}. & & \displaystyle\end{eqnarray}$ (2)

Where K _p0 is additional coefficient of no-load loss; G _Fe is total weight of silicon steel sheet, unit kg; p _Fe is unit mass loss of silicon steel sheet, unit W/kg.

The main internal heat sources of PPEN transformer are load loss of windings and no-load loss of iron core. The magnetic and thermal field indirect coupling calculation scheme can convert the value of total loss to heat production rate (per unit volume) which can be taken as a heat source to solve the temperature field of the PPEN transformer.

3.2. Mathematical model of thermal processes

For oil-immersed power transformer, the dissipation of internal heat is a comprehensive process including heat conduction and heat convection (ignoring radiation) [7].

The heat conduction of iron core and winding is governed by the following form of Poisson equation [8]: $\begin{eqnarray}\displaystyle {\displaystyle \frac{\partial ^{2}t}{\partial x^{2}}}+{\displaystyle \frac{\partial ^{2}t}{\partial y^{2}}}+{\displaystyle \frac{\dot{{\Phi}}}{{\lambda}}}=0. & & \displaystyle\end{eqnarray}$ (3)

Where 𝜆 is the heat conductivity coefficient; $\dot{{\Phi}}$ is the heat source of windings and core (W/m³).

The heat convection of metal surface and oil field meets three basic laws of physics [8], namely conservation of mass, momentum and energy. Which mathematical expressions are as follows:

(1) Mass conservation equation $\begin{eqnarray}\displaystyle \frac{\partial u}{\partial x}+\frac{\partial v}{\partial y}=0. & & \displaystyle\end{eqnarray}$ (4) (2) Momentum conservation equations $\begin{eqnarray}\displaystyle \text{} & \text{} & \displaystyle {\rho}\left(u{\displaystyle \frac{\partial u}{\partial x}}+v{\displaystyle \frac{\partial u}{\partial y}}\right)=F_{x}+{\eta}\left({\displaystyle \frac{\partial ^{2}u}{\partial x^{2}}}+{\displaystyle \frac{\partial ^{2}u}{\partial y^{2}}}\right)\end{eqnarray}$ (5) $\begin{eqnarray}\displaystyle \text{} & \text{} & \displaystyle {\rho}\left(u{\displaystyle \frac{\partial v}{\partial x}}+v{\displaystyle \frac{\partial v}{\partial y}}\right)=F_{y}+{\eta}\left(\frac{\partial ^{2}v}{\partial x^{2}}+{\displaystyle \frac{\partial ^{2}v}{\partial y^{2}}}\right).\end{eqnarray}$ (6)

Where F _x, F _y are the component of force on unit mass of fluid in the direction of x, y, respectively; 𝜂 is dynamic viscosity; 𝜌 is oil density;

(3) Energy conservation equation $\begin{eqnarray}\displaystyle {\rho}c_{p}\left(u{\displaystyle \frac{\partial t}{\partial x}}+v{\displaystyle \frac{\partial t}{\partial y}}\right)-{\lambda}\left({\displaystyle \frac{\partial ^{2}t}{\partial x^{2}}}+{\displaystyle \frac{\partial ^{2}t}{\partial y^{2}}}\right)=\dot{{\Phi}}. & & \displaystyle\end{eqnarray}$ (7)

Where c _p is specific heat capacity.

For isotropic media, equations (3)–(7) are valid under the assumption of a steady-state, as well as the assumption of a homogeneous heat conductivity distribution.

4. Results of simulation

Due to the geometrical complexity of PPEN transformer, the interest is generally focused on the elements of its active part. In this paper, the model is performed using a 2-D modeling, which is comprised of seven active parts by simplifying: iron core (A1), low voltage winding (A2), medium voltage winding (A3), high voltage winding (A4), tap winding (A5), oil volume (A6) and tank (A7). Figure 1 shows the geometry of the model.

Fig. 1.

Geometry of the model.

The geometry of the model is meshed with the smartsize automated grid generation tools of ANSYS workbench. For 2D simple model, this mesh division method is simple and practical, and this method has been fully verified in many practical experience. Figure 2 presents the meshed model. In this paper, the temperature rise of PPEN transformer is simulated based on the maximum loss and the maximum capacity of transformer (300MVA/250MVA/150MVA), the 𝜅 −ϵ turbulence model and SIMPLE algorithm coupling pressure with velocity is used to calculate the temperature field. To solve equations mentioned above by finite element method, some boundary conditions are imposed:

(1) The inlet is mass flow rate boundary;

(2) The outlet is static pressure boundary;

(3) Insulation of winding is set to thin-walled element;

(4) The surface of winding and core is set to fluid-solid coupling;

(5) The temperature of environment is 30 °C.

Fig. 2.

The meshed model.

Among the seven major components of PPEN transformer, the temperature rise of windings (HV/MV/LV) is the greatest since the heat conductivity coefficient of copper is the highest. The temperature rise results of windings are shown in Table 2.

Table 2

Temperature rise results of windings

Item	HV winding	MV winding	LV winding
Winding average temperature rise (K)	50.4	52.7	52.4
Hot-spot of winding (K)	69.1	72	71.4

From Table 2, it has been known that the temperature rise of MV winding is higher than the HV winding and the LV winding. So the detailed results of MV winding are discussed in this paper.

Figure 3 shows the cloud image of temperature distribution of MV winding in the transformer after reaching the steady condition. It can be seen that the temperature rise of the upper section of MV winding is the greatest and the hot-spot is located in the upper part of MV winding.

Figure 4 shows the curve of hot-spot temperature rise, oil temperature average and winding temperature average of MV winding respectively. It can be seen in Fig. 4 that the hot-spot temperature rise of MV winding is 72 K and hot-spot location is in NO.107 disk of MV winding.

Fig. 3.

Cloud image of temperature distribution of MV winding.

Fig. 4.

Temperature rise curve inside MV winding.

5. Temperature rise test of PPEN transformer

5.1. Methods of temperature rise test

The test methods of temperature rise include: direct load method, back-to-back method, circulating current method, zero sequence current method and short-circuit method [9]. The direct load method applies to power transformers with medium and low test capacity. The back-to-back method and circulating current method need two power suppliers, and test connecting systems are relatively complicated. Zero sequence current method is only applicable to some power transformers with bigger section area of neutral point lead and low test capacity. And the short-circuit method is the most common method for oil-immersed power transformers. By a long term practical test experiences in factories, it has been proved that this method can obtain satisfactory measure data in temperature rise test. In this paper, we take this method as the verification method.

5.2. Temperature rise test by short-circuit method

There are five methods of calculating the load loss: H-M (300/300MVA), H-L (150/150MVA), M-L(150/150MVA), H-M-L (300/150/150MVA), H-M-L (300/250/50MVA). Based on calculating results of load loss in different operations, the maximum loss is at H-M-L combined operation (300/150/150MVA). So to validate the feasibility and correctness of the proposed model and method, temperature rise test of this transformer is made at full-load operation (300/150/150MVA). Then experiment data and simulation results are compared.

According to IEC60076-2: 2011, temperature rise test is made by a short-circuit method at the rated frequency. Figure 5 shows the short-circuit test diagram. LV terminals are connected to MV terminals and a three-phase voltage is applied to HV terminals. The power taken by transformer under test is 846.1 kW (the maximum loss) for above ten hours, then the temperature rise of top oil and average oil is measured. The rated current is applied to the winding for one hour, then the hot resistances of windings on the transformer are measured. The hot-spot winding temperature is measured by optical fiber. The optical fiber is mounted on the hot-spot position of winding which is predicted by results of simulation.

Fig. 5.

Short-circuit test diagram.

The measuring results of temperature rise test are recorded in Table 3. During the test, the hot-pot of the tank is located at the busing of the medium voltage winding by scanning the transformer using Infrared Thermography. The picture of the test measurement is shown in Fig. 6.

Fig. 6.

The hot-spot location of tank (ambient temperature: 36 °C).

Table 3

The results of temperature rise test

Item	HV winding	MV winding	LV winding
Winding average temperature rise (K)	45.8	44.1	47.0
Hot-spot of winding (K)	64.5	66.2	62.2
Oil average temperature rise (K)		23.0
Top oil temperature rise (K)		38.2

The simulation value and the test measurement value of temperature rise are compared with allowable limit values of IEC60076-2:2011. The results are shown in Table 4. It shows that all simulated and test measured values of temperature rise are less than the allowable limit values of IEC standard. It is proved that the temperature rise performance of PPEN transformer meets the requirements of ITER organization and IEC60076-2:2011. In addition, the simulation value is bigger than testing measurement value. The reason is that simulation analysis is based on all windings under rated capacity operation (300MVA/250MVA/150MVA), while the temperature test is based on actual full-load operation of transformer (300MVA/150MVA/150MVA). Moreover, the slighly increase of the temperature rise in the simulation is casued by the more severe conditon. Nevertheless, the result demostrates the correctness of the simulation.

Table 4

Simulated/test measured/Allowable limits values of temperature rise

Item		Simulated (K)	Test measured (K)	Allowable limits (K)
Winding average Temperature rise	HV	50.3	45.8	65
	MV	52.7	44.1
	LV	52.4	47.0
Hot-spot of winding	HV	69.1	64.5	78
	MV	72.0	66.2
	LV	71.4	62.2
Oil average temperature rise		27.8	23.0	-
Top oil temperature rise		42.1	38.2	60

6. Conclusion

In this paper, the power loss of transformer is analyzed and calculated, and the value of load loss and no-load loss is acquired; The distribution of temperature field of PPEN transformer is calculated, the winding average temperature rise, top oil temperature rise and hot-spot of winding are obtained from the results of simulation; Meanwhile, the temperature rise test of PPEN transformer has been carried out, the simulation values and testing measurement values are all smaller than the allowable limit values of IEC standard, so it is proved that the temperature rise performance of PPEN transformer meets the requirements of ITER international organization and IEC 60076.

Disclaimer

The view and opinion expressed herein does not necessarily reflect those of the ITER organization.

Footnotes

Acknowledgements

The authors would like to express gratitude to Ministry of Science and Technology of China for the foundation and staff of Institute of Plasma Physics, Chinese Academy of Sciences (ASIPP) for helpful discussions and suggestions.

References

Shimomura

and Spears

, Review of the ITER project, IEEE Trans. Appl. Supercond. 14(2) (2004), 1369–1375.

Gao

Review and analysis of the AC/DC converter of ITER coilpower supply, in: Twenty-Fifth Annual IEEE Applied Power Electronics Conference and Exposition (APEC), 2010, pp. 1810–1816.

Gao

, Preliminary design of the poloidal field AC/DC converter system for ITER coil power supply, Fusion Sci. Technol. 64(4) (2013), 741–747.

Mondino

P.L.

, The ITER pulsed power supply system [C], in: Fusion Engineering, 1997. IEEE/NPSS Symposium, Vol. 1, IEEE, 1997, pp. 491–496.

Yin

K.N.

, The Design Principle of Transformer, Electric Power Press of China, Beijing, 2003.

Mechkov

Tzeneva

Thermal analysis is using 3D FEM model of oil-immersed distribution transformer, in: 19th International Symposium on Electrical Apparatus and Technologies (SIELA), 2016, pp. 1–3.

Jing

Yuan

Calculation and Analysis of the hot-spot temperature rise for large power transformer based on 3D electromagnetic-fluid-thermal couping method, in: 20th International Conference on Electrical Machines and Systems (ICEMS), 2017, pp. 1–4.

Yang

S.M.

and Tao

W.S.

, Heat Transfer, Higher Education Press, Beijing, 2006, p. 8.

Xie

S.C.

, Power Transformer Manual, China Machine Press, Beijing, 2003, p. 1.

Rated power (MVA)	210/175/105 (ONAN)
	300/250/150 (ONAF)
Rated voltage (kV)	400 ± 8 × 1.25%/66/22.5
Vector group	YNyn0d11
Frequency (Hz)	50
Insulation type	oil-immersed type
Short-circuit Resistance	HV-MV11%, HV-LV25%, MV-LV14%
Temperature rise	60/65/78K(top oil/winding average/hot-spot)
Type of cooling	ONAN/ONAF