Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.771368
Title: Experimental study of transformer liquid flow and temperature distributions
Author: Daghrah, Muhammad
ISNI:       0000 0004 7657 7156
Awarding Body: University of Manchester
Current Institution: University of Manchester
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Determination of the temperature distribution and the so called hot spot temperature within the transformer winding is crucial for both thermal design in the factory and thermal rating during the operation. In oil immersed power transformers, oil acts as the coolant in addition to its role of insulation. The oil circulates within the winding either naturally in Oil Natural (ON) cooling modes or forced to circulate in a zig-zag fashion in Oil Directed (OD) cooling modes. Temperature distribution is heavily dependent on the oil flow distribution. This thesis aims to experimentally investigate the effects of a wide range of operational and geometrical parameters on the transformer oil flow and temperature distributions. An experimental setup based on disc type transformer winding models was established where winding model geometries such as radial cooling duct height, axial cooling duct width, and number of discs per pass can be adjusted and operational conditions such as winding inlet oil velocity, winding inlet oil temperature, and electrical loss distribution within the winding model can be varied. Thermocouple arrays were used to measure temperature distribution. A Particle Image Velocimetry (PIV) system was used to capture the oil flow distribution including the detailed phenomenon of reverse flow. Under OD cooling modes, experimental validations were conducted under isothermal conditions for the application of dimensional analysis on oil flow proportion in and pressure drop coefficient over the winding model. It was verified that if the dimensionless controlling parameters, pass inlet Reynold number and the ratio of radial duct height over axial duct width, are matched, both oil flow proportion and pressure drop coefficient are matched. Under non-isothermal OD cooling modes, it was found that the existence of electrical losses does not affect the oil flow distribution. In addition, extra high pass inlet oil velocity causes oil reverse flow and oil stagnation to occur and hence the hot spot temperature is not necessarily reduced with further increase of inlet oil velocity. Under ON cooling modes, more distorted flow distribution which leads to oil stagnation and hence higher hot spot temperature were observed under higher loading levels or lower winding inlet oil velocities. Inlet oil temperature and non-uniform losses in the winding model showed minor impact on the oil flow distribution. In terms of geometric parameters, higher radial cooling duct or higher number of discs per pass makes oil stagnation easier to occur which could significantly increase the hot spot temperature. Thermal performances of conventional mineral oil and alternative liquids including gas-to-liquid oil and synthetic ester oil were compared using the zig-zag disc type winding model under both OD and ON cooling modes. Mineral oil and gas-to-liquid oil showed similar behaviours under both cooling modes. Under OD cooling conditions, the synthetic ester is more resistant to oil reverse flow due to its high viscosity, which leads to more uniform oil flow distributions and hence lower hot spot temperature but with the cost of higher pressure loss. Under ON cooling modes, retro-filling scenarios were investigated and it was shown that synthetic ester causes lower inlet flow rates due to its higher viscosity and hence increases the hot spot temperature compared to the other oils.
Supervisor: Wang, Zhongdong ; Liu, Qiang Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.771368  DOI: Not available
Keywords: Particle Image Velocimetry ; Experiment ; CFD Modelling ; Power Transformers ; Thermal Modelling
Share: