Investigating temperature rise dynamics at hot-spots within dry-Type transformer windings: A comparative analysis across varied loading rates and an extrapolative computational model

Abstract

The dynamic temperature data of a transformer can be utilized to accurately assess its load capability and operational longevity. This paper presents a study on the temperature rise characteristics of hot-spots in dry-type transformer windings and develops a extrapolative computational model based on various load rates, utilizing the results of transformer temperature rise tests. Firstly, the dynamic temperature rise characteristics of B-phase low-voltage (LV) winding and high-voltage (HV) winding hot-spot, along with its correlation calculation model with varying load rates, were determined based on the results of temperature rise tests. Subsequently, an extrapolative computational model for the temperature rise of the winding hot-spot, utilizing a three-section continuous load application approach, was proposed employing the lumped parameter method. The validity of this model was confirmed through verification. Lastly, a temperature rise test was conducted on another dry-type transformer to validate the universality of the model. Finally, a different dry-type transformer is chosen, and a temperature rise test is conducted to confirm the general applicability of the model. The results indicate that the calculation model for the hot-spot temperature rise of the winding under three consecutive loads is associated with the dynamic temperature rise pattern of the hot-spot of the winding between the temperature rise test with a small load rate and the temperature rise test with a large load rate. The maximum deviation between the hot-spot temperature rise test of the winding of SCB14-800/10-NX2 transformer and the results of the steady-state temperature rise value calculation is only 1.3 K, with a maximum error of the time constant value of the temperature rise being only 0.05 h. Similarly, the maximum deviation of the steady-state temperature rise value obtained from the hot-spot temperature rise test and the extrapolative computation of SCB-1250kVA/10 kV transformer winding is only 2.03 K, with a maximum error of the temperature rise time constant value being only 0.09 h. The extrapolative computational model proposed in this study conducts the temperature rise test under a small load rate instead of a large load rate, addressing the industry challenge of the challenging implementation of transformer on-site temperature rise tests and high power loss. This approach enhances inspection efficiency.