Study on the Airflow and Convective Heat Transfer of High-Voltage Transformers

The total power dissipation of the radiator equals the total power dissipation of the high-voltage transformer minus the heat loss of the transformer oil tank. The oil temperature at the bottom of the transformer is taken as the average of the oil temperature at the radiator outlet and the oil temperature at the high-voltage transformer tank. This value serves as the starting point for calculating the change in coil temperature rise, so variations in the bottom oil temperature directly affect the coil temperature. Since the primary function of the radiator is to lower the bottom oil temperature of the transformer, the efficiency of the radiator has a significant impact on the temperature rise of the transformer coils. Airflow resistance, pressure, and the pressure difference between the radiator inlet and outlet are the main factors determining the air velocity through the radiator.

Temperature rise tests were conducted on a 750 MVA transformer at the LINDOM substation, where the transformer was asymmetrically placed in a transformer room without a roof. The transformer used eight radiators arranged in two rows, employing forced oil circulation with air-blown cooling. The power loss of the load is approximately 2 MW. To study the airflow and convective heat transfer around the high-voltage transformer, the following three test schemes were adopted:

(1) Radiator 4 m from the wall, to study the impact of the wall on airflow around the transformer in this space.

(2) Radiator 2 m from the wall, to study the impact of the wall on airflow around the transformer in a reduced space.

(3) Radiator 2 m from the wall, with skirts added below the radiator to enhance heat dissipation and prevent air recirculation.

Calculations were performed for the three schemes, and measurements were taken under normal operating conditions at 3-hour intervals. In scheme 1, although the radiator had no skirt, it was far from the wall, resulting in a low temperature rise (37 K). In scheme 2, the radiator had no skirt and was closer to the wall, causing the bottom oil temperature to rise (45 K). In scheme 3, adding skirts below the radiator significantly reduced the temperature rise (39 K). This result indicates that increasing the transformer room area as in scheme 1 can effectively reduce the bottom oil temperature of the transformer. Adding skirts below the radiator promotes more uniform air distribution in the transformer room, increases the utilization of the radiator, and also lowers the oil temperature.

In scheme 2, the ambient air temperature rise (13 K) was lower than that in scheme 3 (15 K), but the bottom oil temperature rise in scheme 2 (45 K) was higher than in scheme 3 (39 K). This is because, under constant transformer power loss and unchanged transformer room layout, for scheme 2, the heat generated by the transformer coil and other heating elements accumulates in the oil and cannot be effectively dissipated into the air without skirts, resulting in a higher bottom oil temperature but a lower ambient air temperature rise. For scheme 3, the addition of skirts improved the radiator efficiency, significantly lowering the transformer oil temperature, while the extra energy dissipated into the environment led to a noticeable increase in ambient air temperature. Therefore, scheme 3 balances the transformer and ambient air temperatures to achieve a lower oil temperature, representing an economical and efficient method.