Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Current signal processing-based techniques for transformer protection

Etumi, Adel 2016. Current signal processing-based techniques for transformer protection. PhD Thesis, Cardiff University.
Item availability restricted.

[thumbnail of 2016EtumiAPhD.pdf]
Preview
PDF - Accepted Post-Print Version
Download (6MB) | Preview
[thumbnail of EtumiA.pdf] PDF - Supplemental Material
Restricted to Repository staff only

Download (200kB)

Abstract

Transformer is an expensive device and one of the most important parts in a power system. Internal faults can cause a transformer to fail and thus, it is necessary for it to be protected from these faults. Protection doesn’t mean that it prevents damage to the protected transformer but it is to minimize the damage to the transformer as much as possible, which consequently minimizes the subsequent outage time and repair cost. Therefore, fast and reliable protection system should be used for limiting damages to the transformer by rapidly disconnecting the faulty transformer from the network, which also leads to the elimination of the stresses on the system itself and preventing damage to adjacent equipment. The main aim of this thesis is to propose transformer protection technique that is fast and highly sensitive to internal faults that occur inside the transformer, to overcome the problems of current transformer saturation and inrush current, and to make it immune to the external faults (through faults) that occur outside of the transformer protection zone. The current transformer saturation and inrush current are significant problems since they cause malfunction of the protection system, which consequently will disconnect the transformer because they are considered faults. This improper disconnection of transformer is not desirable as it shortens its life time. So the proposed protection technique was designed to be fast and to avoid maloperation caused by saturation and inrush current. The proposed protection technique was based on current signal processing. Three methods, namely the application of correlation coefficients, current change ratio (CCR) and percentage area difference (PAD) were proposed based on practical and simulation tests. These techniques were successfully proved by carrying out tests on Simulink models using MATLAB/SIMULINK program and on a practical laboratory model. In transformer transient state, the response time for the methods that were used to address the problem of inrush condition, was 10 ms for CCR when transformer was on no-load and 5 ms for PAD when the transformer was on-load. This response time Current Signal Processing-Based Techniques for Transformer Protection v is faster than the most popular method relying on second harmonic, which needs at least one cycle (20ms in 50 Hz systems) to recognize the condition. In transformer steady state, it was proved that the proposed correlation method was capable of detecting the internal faults successfully within a very short time, ranging from 0.8 to 2.5 ms according to the type and severity of the fault and in addition was able to overcome the problem of current transformer (CT) saturation. The contribution of this research is the development of a transformer protection technique, which is simple in design, fast and reliable in fault detection and at the same time capable of overcoming the problems of current transformer saturation and inrush current.

Item Type: Thesis (PhD)
Date Type: Publication
Status: Unpublished
Schools: Engineering
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Uncontrolled Keywords: Transformer protection; Current transformer saturation; Inrush current problem; Transformer failure; Differential relay problem; Correlation coefficient concept.
Date of First Compliant Deposit: 22 September 2016
Last Modified: 28 Sep 2021 14:44
URI: https://orca.cardiff.ac.uk/id/eprint/94716

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics