There has been significant development in the area of neural network based power system protection in the previous decade. Neural network technology has been applied for various protective relaying functions including distance protection. The reliability and efficiency of ANN based distance relays is improving with the developing digital technologies. There are, however, some inherent deficiencies that still exist in the way these relays are designed. This research addresses some of these issues and proposes an improved protective relaying scheme.<p> The traditional ANN distance relay designs use parameter estimation algorithms to determine the phasors of currents and voltages. These phasors are used as inputs to determine the distance of a fault from relay location. The relays are trained and tested on this criterion; however, no specific relay characteristic has been defined. There is a need for development of a new methodology that will enable designing of an ANN that works as a generic distance relay with clearly defined operating boundary.<p> This research work presents a modified distance relaying algorithm that has been combined with a neural network approach to eliminate the use of phasors. The neural network is trained to recognize faults on basis of a specific relay characteristic. The algorithm is flexible and has been extended for the design of other relays. The neural network has been trained using pure sinusoidal values and has been tested on a 17-bus power system simulated in PSCAD. The training and testing of the neural network on different systems ensures that the relay is generic in nature. The proposed relay can be used on any transmission line without re-training the neural network.<p> The design has been tested for different fault conditions including different fault resistances and fault inception angles. The test results show that the relay is able to detect faults in lesser time as compared to conventional relay algorithms while maintaining the integrity of relay boundaries.

There has been significant development in the area of neural network based power system protection in the previous decade. Neural network technology has been applied for various protective relaying functions including distance protection. The reliability and efficiency of ANN based distance relays is improving with the developing digital technologies. There are, however, some inherent deficiencies that still exist in the way these relays are designed. This research addresses some of these issues and proposes an improved protective relaying scheme.<p> The traditional ANN distance relay designs use parameter estimation algorithms to determine the phasors of currents and voltages. These phasors are used as inputs to determine the distance of a fault from relay location. The relays are trained and tested on this criterion; however, no specific relay characteristic has been defined. There is a need for development of a new methodology that will enable designing of an ANN that works as a generic distance relay with clearly defined operating boundary.<p> This research work presents a modified distance relaying algorithm that has been combined with a neural network approach to eliminate the use of phasors. The neural network is trained to recognize faults on basis of a specific relay characteristic. The algorithm is flexible and has been extended for the design of other relays. The neural network has been trained using pure sinusoidal values and has been tested on a 17-bus power system simulated in PSCAD. The training and testing of the neural network on different systems ensures that the relay is generic in nature. The proposed relay can be used on any transmission line without re-training the neural network.<p> The design has been tested for different fault conditions including different fault resistances and fault inception angles. The test results show that the relay is able to detect faults in lesser time as compared to conventional relay algorithms while maintaining the integrity of relay boundaries.

This work presents a new Loss-of-Excitation in synchronous generators protection scheme based on the Continuous Apparent Impedance Calculation. This methodology aiming to improve security and dependability to the ANSI 40 function in order to distinguish cases of loss of excitation from other external disturbances such as stable power swings caused by the elimination of short circuits, for instance. Traditional loss of excitation protection methods may operate incorrectly in the occurrence of this kind of disturbance. Since the proposed methodology have settings related to the reactance parameters of the protected generator (Xd e X d), tests have been carried out considering two different machines with distinct parameters of the WSCC 9-bus system. Aiming to test the proposed technique and compare it with other widely used methods Mason, Berdy and Positive Offset, three kinds of events were simulated, namely, total loss of excitation, partial loss of excitation and stable power swing, for different initial generator loading. The computer simulation results demonstrate that the proposed method correctly identifies the occurrence of loss-of-excitation and does not operate in situations of stable power swings, regardless of the parameters of generators and their operating condition. Its alarm function allows taking the underexcited machine out faster by using manual intervention, which is especially useful when the initial loading is light or when it‟s a partial loss of excitation case. Another highlight is the ease adjustment of the method, which is related only with the generator‟s reactances data. / Este trabalho apresenta um novo método de proteção contra perda de excitação de geradores síncronos fundamentado no Cálculo Contínuo da Impedância Aparente. Este novo conceito visa garantir segurança e disponibilidade para a função de proteção ANSI 40 através da distinção de casos de perda de excitação e outros distúrbios externos ao gerador, tais quais as oscilações estáveis de potência ocorridas após a eliminação de curto-circuitos. As técnicas tradicionais de proteção contra perda de excitação podem apresentar operações indevidas quando da ocorrência de tais distúrbios. Considerando que a metodologia desenvolvida possui ajustes em função dos parâmetros das reatâncias do gerador protegido (Xd e X d), são feitos testes com duas máquinas de parâmetros distintos do sistema WSCC 9-barras. Visando testar a técnica proposta e compará-la com métodos amplamente utilizados Mason, Berdy e Offset Positivo, foram simulados três tipos de eventos, ou seja, perda de excitação total, perda de excitação parcial e oscilação estável de potência, para diferentes pontos operacionais do gerador. Os resultados das simulações computacionais demonstraram que o método proposto identifica corretamente a ocorrência de perda de excitação e não opera em situações de oscilação estável de potência, independentemente dos parâmetros do gerador e de sua condição de operação. A função de alarme permite desligar a máquina subexcitada mais rapidamente através de intervenção manual, o que é especialmente útil quando o carregamento inicial é baixo ou a perda de excitação é parcial. Destaca-se ainda a facilidade de ajuste do método, o qual depende apenas dos dados das reatâncias do gerador.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This work aims to evaluate the performance of the loss of field protection methods in synchronous generators. The methods are introduced to make available in a single text the various ways to protect the synchronous generator against the loss of field. The conventional methods have some problems, and the main of them is the improper operation caused by the power stable oscillations. In order to solve or minimize the relay algorithm malfunction, two new methods of loss of field protection are proposed. The first is designed to increase the operational area of the generator on steady state conditions, bounded by the conventional loss of field protection. It was accomplished by modifying the operational characteristic of the mho relay, which is better coordinated with the generator capability curve. The second makes use of loss of field conception in a fuzzy set theory. With the objective to identify the performance of each one, the methods were evaluated through computational simulations of loss of field and stable power oscillation. The methods were set and evaluated according to the generator parameters (Xd and X'd). Since, tests in three machines with different parameters were carried out. So it was possible to conclude that the methods do not behave the same way for different generators parameters. On the other hand, the proposed technique, which is based on the fuzzy set theory was more efficient and not have been affected by the generator parameters and system considered. / Este trabalho visa avaliar o desempenho dos métodos de proteção contra a perda de excitação em geradores síncronos. Os métodos são apresentados de forma a tornar disponível em um único texto as diversas maneiras de se proteger o gerador síncrono contra a perda do seu sistema de excitação. Os métodos convencionais abordados apresentam alguns problemas, sendo o principal, a operação indevida causada por oscilações estáveis de potência. Visando solucionar ou minimizar estes problemas, dois novos métodos de proteção contra a perda de excitação são propostos. O primeiro, objetivando aumentar a área operacional do gerador em regime permanente, restringida pela proteção contra a perda de excitação convencional, por meio de uma característica operacional modificada, melhor coordenada com a curva de capacidade do gerador. O segundo introduz os fundamentos clássicos da proteção contra a perda de excitação na teoria dos conjuntos nebulosos. Com o objetivo de se identificar o desempenho de cada um, os métodos existentes e os propostos foram avaliados por meio de simulações computacionais de perda de excitação e oscilação estável de potência. Como os métodos possuem ajustes em função dos parâmetros do gerador protegido (Xd e X d), realizaram-se testes com três máquinas de parâmetros distintos. Deste modo, foi possível concluir que os métodos não se comportam da mesma maneira para geradores de diferente porte. Por outro lado, a técnica proposta, que tem como base a teoria dos conjuntos nebulosos se mostrou eficiente e não teve seu desempenho afetado pelos parâmetros do gerador e do sistema considerado.