High voltage transmission system availability and system security are key performance criteria for electricity utilities worldwide. System disturbances need to be cleared quickly and accurately in order to minimise the impact of faults and to facilitate speedy system restoration. In this context, the South African utility, Eskom has maintained a process of refreshing protective relaying technology as older equipment becomes obsolete and is no longer capable of meeting the utility’s requirements. The difficulties which a process of equipment renewal presents the organisation with include the risk of incorrectly applying the newer technologies within the complex electrical network. The application of new technology is affected by the complexities of the newer technology with respect to the older, more familiar technologies. Some of the difficulties can be addressed with revised commissioning procedures or the use of modern test equipment. Enhanced relay algorithms and settings calculation methodologies can however not be simplified. Protective relay maloperations cannot always be completely avoided and when they do occur, these must be investigated and addressed to prevent future recurrences. The research covered by this dissertation focuses on a number of protective relay maloperations on transmission lines using impedance protection algorithms. The research undertaken identifies the previously unidentified causes of the maloperations and describes a relay settings solution for improving the accuracy of the protective relays. The methodology that was followed in the research covers the following aspects: • Identification and highlighting of some of the protection relay maloperations that occurred during system faults, • Review of the fundamental principles involved in system fault analysis, • Comprehensive study of the theory involved in the calculation of an overhead line conductor self and mutual-inductance, as well as the calculation of the positive, negative and zero sequence impedances of an overhead line, • Brief evaluation of the effect of load impedance on relay measurements and the impact on fault clearing operation, • Analysis of the theoretical operation of various numerical relays during singlephase- to-earth faults in radial and meshed (complex) network conditions, • Mathematical calculations using typical Newton-Raphson methods to study the impact of resistive single-phase-to-earth faults on the voltage and current measurements at the relaying position with the exclusion of the capacitive components between conductors and conductors and earth, • Comparison and evaluation of mathematical calculations and system studies using network simulation software which included all steady state network parameters, • Review and analysis of actual system faults that had been previously analysed without definitive conclusion. The faults were re-analysed in an attempt to correlate findings with the hypothesis of the research, • Comparison of the performance of protective relay impedance charactersitics using positive sequence domain versus loop domain analysis techniques. This study concluded that significant benefits can be achieved by analysing system faults and relay operation using loop quantities in primary impedance values as opposed to positive sequence or apparent impedance quantities in secondary values. The inherent differences between the positive or apparent impedance characteristics of the relays are nullified when considered in the loop impedance domain, provided that the relays reach settings were calculated correctly. The study also showed that load current cannot be ignored when calculating settings as it has significant impact on the actual impedance measured during fault conditions. It is therefore crucial that when relays from different manufacturers are being used to protect the same circuit that the differences between the relays and the subsequent measurements are clearly understood and compensated for. Finally relay setting changes have been proposed for implementation based on the findings of this research. The combination of the theory, network simulations and secondary injections performed on the relays all correlate and therefore validate the research. It is left for the utility and or users of these relays to evaluate the results of this research and implement the necessary changes as applicable. / Thesis (MSc (Electrical Engineering))--North-West University, Potchefstroom Campus, 2012

High voltage transmission system availability and system security are key performance criteria for electricity utilities worldwide. System disturbances need to be cleared quickly and accurately in order to minimise the impact of faults and to facilitate speedy system restoration. In this context, the South African utility, Eskom has maintained a process of refreshing protective relaying technology as older equipment becomes obsolete and is no longer capable of meeting the utility’s requirements. The difficulties which a process of equipment renewal presents the organisation with include the risk of incorrectly applying the newer technologies within the complex electrical network. The application of new technology is affected by the complexities of the newer technology with respect to the older, more familiar technologies. Some of the difficulties can be addressed with revised commissioning procedures or the use of modern test equipment. Enhanced relay algorithms and settings calculation methodologies can however not be simplified. Protective relay maloperations cannot always be completely avoided and when they do occur, these must be investigated and addressed to prevent future recurrences. The research covered by this dissertation focuses on a number of protective relay maloperations on transmission lines using impedance protection algorithms. The research undertaken identifies the previously unidentified causes of the maloperations and describes a relay settings solution for improving the accuracy of the protective relays. The methodology that was followed in the research covers the following aspects: • Identification and highlighting of some of the protection relay maloperations that occurred during system faults, • Review of the fundamental principles involved in system fault analysis, • Comprehensive study of the theory involved in the calculation of an overhead line conductor self and mutual-inductance, as well as the calculation of the positive, negative and zero sequence impedances of an overhead line, • Brief evaluation of the effect of load impedance on relay measurements and the impact on fault clearing operation, • Analysis of the theoretical operation of various numerical relays during singlephase- to-earth faults in radial and meshed (complex) network conditions, • Mathematical calculations using typical Newton-Raphson methods to study the impact of resistive single-phase-to-earth faults on the voltage and current measurements at the relaying position with the exclusion of the capacitive components between conductors and conductors and earth, • Comparison and evaluation of mathematical calculations and system studies using network simulation software which included all steady state network parameters, • Review and analysis of actual system faults that had been previously analysed without definitive conclusion. The faults were re-analysed in an attempt to correlate findings with the hypothesis of the research, • Comparison of the performance of protective relay impedance charactersitics using positive sequence domain versus loop domain analysis techniques. This study concluded that significant benefits can be achieved by analysing system faults and relay operation using loop quantities in primary impedance values as opposed to positive sequence or apparent impedance quantities in secondary values. The inherent differences between the positive or apparent impedance characteristics of the relays are nullified when considered in the loop impedance domain, provided that the relays reach settings were calculated correctly. The study also showed that load current cannot be ignored when calculating settings as it has significant impact on the actual impedance measured during fault conditions. It is therefore crucial that when relays from different manufacturers are being used to protect the same circuit that the differences between the relays and the subsequent measurements are clearly understood and compensated for. Finally relay setting changes have been proposed for implementation based on the findings of this research. The combination of the theory, network simulations and secondary injections performed on the relays all correlate and therefore validate the research. It is left for the utility and or users of these relays to evaluate the results of this research and implement the necessary changes as applicable. / Thesis (MSc (Electrical Engineering))--North-West University, Potchefstroom Campus, 2012