Development of the power gridtowards a more reliable and smarter system requires frequent on-line monitoring of the power components. Power transformers and their bushings are particularly important components in a power transmission system and their insulation degradation may lead to catastrophic failures. Time consuming and costly replacement of these components raise the importance of their frequent monitoring. A fault in a power transformer bushing can also involve in the failure of the transformer. Therefore, on-line diagnostics of power transformers and their bushings is of great interest. Several methods exist for diagnostics of these components. However, some of them can only be done off-line in maintenance periods, and the existing on-line methods generally provide less information, especially on the internal solid insulation parts. In this project, a new technique for on-line diagnostics of the power transformer and the bushing insulation is proposed. In this technique, natural transients happening in the power system such as lightning and switching surges can be used as stimuli for on-line dielectric response measurements. This technique can provide information on insulation close to what Dielectric Spectroscopy offers in off-line measurements. The wide-ranging frequency content of power system transients is their advantage for being usedas stimuli when measuring the Dielectric Response. The response can have particular signatures due to different types of defects in the insulation varying with frequency. Oil-impregnated paper as a major insulation component in power transformer and its bushing has been investigated in this project. Moisture content and temperature, as two important degradation factors in this type of insulation, have been studied to evaluate the performance of the proposed technique in the diagnostics of the oil-impregnated paper. The results are verified with the dielectric response obtained through commercial instruments. The results show that the proposed technique has the ability to track the changes in dielectric response due to the moisture content and temperature. Measurements were done at both highvoltage (40kV) and low voltage (10V) levels, and the corresponding circuit models to achieve reasonable accuracy for the results are discussed. Moving on from the material samples, a further study was done on three service-aged 150 kV bushings to investigate the feasibility of the technique on the diagnostics of power transformer bushings. Their dielectric response measured by the transient stimuli showed good agreement with their response obtained by the commercial instruments. The effect of the transformer winding on the transient response of the bushing is a further aspect of the real conditions for on-line diagnostics. This has been investigated through the simulation of transient models for transformers and bushings, and possible solutions for distinguishing the responses are presented. The proposed new on-line diagnostics technique by utilizing natural transients can provide information about the insulation system in a certain range of frequency without interrupting the operation or requiring an external voltage source. However, the validity range of the results depends on the bandwidth of the applied transients and other measurement considerations. This approach can be valuable in frequent monitoring of dielectric properties of the power transformers and their bushings as a complement to the other available on-line techniques. / <p>QC 20140409</p>
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kth-144092 |
| Date | January 2014 |
| Creators | Nikjoo, Roya |
| Publisher | KTH, Elektroteknisk teori och konstruktion, Stockholm |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Licentiate thesis, comprehensive summary, info:eu-repo/semantics/masterThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | TRITA-EE, 1653-5146 ; 2014:018 |
Page generated in 0.0015 seconds