M.Ing. (Electrical and Electronics) / The increased use of power electronic equipment in power networks prompted the development of various topologies to compensate for the distortion in the power networks. The various compensator topologies employ a vast range of converters for the compensation of the different non-active power components. The compensators are either designed to eliminate a specific non-active power component, or a combination of converters is used to simultaneously compensate for several non-active components. The choice of compensator depends largely on the type of load, the distortion levels in the power network, the effectiveness of the compensator and very importantly, the cost implications for the user. Under constant load conditions a particular compensator would suffice. It is however not the case when the load and the accompanied distortion varies with time, which is the case with present non-linear, dynamic high power loads on the network. In these cases,. a need for another compensator or compensation strategy, that is more effective in compensating the changing load condition, exists. It would therefore be advantageous to construct a single compensator from various converters -the hybrid compensator -, so as to enable the user to compensate effectively at all times the distortion caused by his load. In order to be able to operate such a hybrid compensator cost-effectively an intelligent control system capable of constantly monitoring the load and updating the compensation strategy, is needed. Keeping in mind that, with the technology available today, compensators can effectively operate for periods in excess of twenty years, it makes sound economical sense to operate the compensator as cost-optimally as possible. This dissertation investigates the development of an artificial neural network based controller for the cost-optimal control of a hybrid compensator. The hybrid compensator considered consists of the following: A 21 kVAR three phase FF-TCR compensator with LC-fiIters tuned at the 5th, 7th,11th and 13th harmonic frequencies and a 6 kVA three-phase dynamic power filter. The hybrid compensator is to be applied for the compensation of a 25 KVA non-linear load (Inductively loaded controlled rectifier). The above mentioned compensators have been modelled to agree with experimental pilot plants. The complete system with low-level controllers was simulated with EMTP (The Electromagnetic Transients Program). This simulation was used to verify the intelligent controller operation. The neural network based controller that is investigated, consists of a Backpropagation-trained neural network, that continuously analyses the load conditions, considers the operational characteristics and losses of the hybrid compensator and proposes a cost-optimal compensation strategy for the hybrid compensator. The modelling of the hybrid compensator's operational losses and characteristics to enable the cost-effective operation thereof is discussed. Special attention is given to the modelling of the cost-effective control strategy, in the training data used for the training of the neural network controller. The training of the neural network controller, and an evaluation of its behaviour when applied to two different hybrid compensator structures, is also given.

M.Ing. (Electrical and Electronic Engineering) / Shunt reactors are used in power system amongst other things, to suppress overvoltages in the network during network switching, auto reclosing of transmission lines and under low loading condition of the network. Traditionally, shunt reactors of fixed type have been used and these have been permanently connected to the system. This research work investigated the feasibility of replacing the fixed shunt line reactors with a reactor, whose parameters are automatically varied depending on the system reactive power requirements, in a mature power system overlaid with high voltage lines to improve amongst, other things voltage stability. However, in a mature system overlaid with parallel higher voltage lines, power tend to flow on the matured system rather than higher voltage system as predetermined by various impedances of the power system. It is therefore desirable that loading of higher voltage lines be increased because of their higher power transfer capability and fact that higher voltage networks generate capacitive power which is substantially more than for each line at the original voltage. However, replacing a large number of fixed line reactors, at low loadings of higher voltage lines, even when system collapse is averted by increasing the number of reactors inserted into the system, overvoltage problems may still be an issue, until a certain number of must-run reactors, operating at full reactance, are put into service. If too much inductive reactance is removed from the system when the loading levels are extremely low, the power system will collapse. Therefore, there was no economic benefit in replacing all the fixed shunt line reactors with controllable type when the loading was still relatively low. Thus, the majority of the converted reactors were operating in their full rating as there was still more reactive power to be absorbed. However, when the power was diverted to flow on the higher voltage system through the use of series compensation of the higher voltage system, there was a possibility of making some fixed shunt line reactors on the higher voltage network controllable. This also allowed more power to be transferred in the higher voltage system thus improving its utilisation. Also, a positive impact on active system losses was realised.

The main objective of this study is to develop a methodology to assist in the reconfiguration and expansion of an electrical sub-transmission network within the framework of Value-based planning. This methodology applies to typical municipal networks found in South Africa. A literature study indicates that most of the fundamental concepts for power system reliability is well established with extensive research done in North America, and other parts of the world. Reliability worth assessment of sub-transmission networks in South Africa, which include composite networks and substation reliability, is however not well developed. From a Value-¬based point of view the literature further does not provide much linkage to the evaluation of alternative long-term network options. This is especially true in terms of the life cycle cost assessment of network alternatives, which include the prediction of customer damages as a function of network performance. In this dissertation a methodology is proposed which utilises the basic network reliability concepts to assess the performance of existing and future alternative network options. The load point Expected Unserved Energy is used to quantify network performance and is obtained through a contingency enumeration process. An existing Geographical Load Forecasting technique defines all customers connected to a load point, on a homogeneous level. This information along with Sector Customer Damage Functions is used to predict existing and future Composite Customer Damage Functions at the associated load point in the sub-transmission network. To arrive at the total minimum cost, which is the objective of Value-based planning, the present worth for each alternative is obtained from the annual utility and customer cost over the life cycle of the alternative. The alternative that result in the lowest present worth is identified as the preferred alternative. A case study is conducted on the sub-transmission network of the Greater Pretoria Metropolitan Council (GPMC) in order to prove the methodology. The entire network is analysed in order to identify the sub-system with the worst performance from a reliability point of view. Alternative network options are identified and the methodology is used for the evaluation of these alternatives. The application of this methodology provides the network planner with the ability to make better decisions with regard to the allocation of reliability. Through the calculation of reliability indices, tangible guidelines can be provided to quantitatively assess the impact of different network alternatives. These guidelines assess contingency probabilities explicitly and along with reliability worth evaluation provide a fundamental tool to conduct Value-based planning. The application of this methodology can lead to significant savings in capital investment while maintaining an acceptable level of reliability. / Dissertation (M Eng (Electrical Engineering))--University of Pretoria, 2006. / Electrical, Electronic and Computer Engineering / unrestricted

This research is aimed at analyzing the performance of the restricted earth fault relay used on a 400kV capacitor bank scheme used on the Eskom Transmission network. After the commissioning of two Capacitor Banks using the above mentioned scheme design the plant was energized. As a result of the energization the Shunt Capacitor Bank (SCB) then experienced spurious trips. The cause of the trips was found to be the operation of the restricted earth fault relay. In this research project, restricted earth fault protection on SCBs in the Eskom Transmission environment will be the area of interest. The proposed research will analyze the current SCB protection scheme in service specifically looking at the restricted earth fault circuit design and relay performance. An equivalent model of the SCB from primary plant perspective based on theory will be derived. Recording and analyzing of Comtrade transient waveforms respectively, when the bank is nergized, will be done in order to provide a reference base to work from. Manual alculations of various parameters from the derived model including transient inrush currents and fault currents will be performed to access applicable scheme parameters. Further calculations will include the voltage setting for the restricted earth fault relay. As a result of the analysis a recommendation will be made on a viable solution or a revised design will be put forward, based on the results, to improve the scheme’s performance. The protection scheme is used on the two 400kV 100MVAR capacitor banks 11 and 12 at Hydra Substation.

Transient stability analysis is an important part of power planning and operation. For large power systems, such analysis is very time consuming and expensive. Therefore, an online transient stability assessment will be required as these large power systems are operated close to their maximum limits. In this thesis swallowtail catastrophe is used to determine the transient stability regions. The bifurcation set represents the transient stability region in terms of power system transient parameters bounded by the transient stability limits. The system modelling is generalized in such, that the analysis could handle either one or any number of critical machines. This generalized model is then tested on a three-machine as well as a seven-machine system. The results of the stability analysis done with the generalized method is compared with the time solution and the results were satisfactory. The transient stability regions determined are valid for any changes in loading conditions and fault location. This method is a good candidate for on-line assessment of transient stability of power systems. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

This work demonstrates the use of digital simulation for analyzing protection system performance. For studies of complex, multi-relay protection systems, digital simulation provides utility engineers with an attractive alternative to relay testing techniques. The cost of digital simulation facilities can be lower than the cost of comparable testing facilities; relay hardware does not have to be made available for the test laboratory. Digital simulation would ordinarily be impractical for security and dependability studies, due to the thousands of individual simulations involved. The number of simulations needed can be greatly reduced by using a technique called "numerical logic replacement" for implementing the protection scheme logic. This unconventional technique makes near-misoperation visible from individual simulations. The likelihood of overlooking potential misoperation is thus much lower than with the usual direct (Boolean) implementations. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

Orientadores: Carlos Alberto Castro Júnior, Manfred Fritz Bedriñana Aronés / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-24T09:53:50Z (GMT). No. of bitstreams: 1 Xavier_CarlosEduardo_M.pdf: 2536219 bytes, checksum: cdaf7cc5b96c9695e965ceacd285daa5 (MD5) Previous issue date: 2013 / Resumo: Esta dissertação de mestrado apresenta um método prático para o cálculo do ponto de máximo carregamento (PMC) de sistemas elétricos de potência baseado nas propriedades geométricas do problema, e na utilização de um fluxo de carga de Newton-Raphson com otimização de passo (FCOP). O PMC é obtido considerando uma margem de erro aceitável ?%, sendo esta margem uma informação prática que pode ser definida pelo operador da rede. É também realizado um equacionamento apropriado para aplicação do método prático em áreas previamente definidas. São apresentados resultados de simulações para sistemas de transmissão teste do IEEE, sistemas de distribuição e sistemas reais incluindo o Sistema Interligado Nacional (SIN) com mais de 4.500 barras. Foram realizadas inúmeras simulações com o método proposto a fim de avaliar a sua robustez, tendo sido estipuladas margens de erro aceitáveis de 0,1% até 10% e carregamento inicial para as redes partindo do caso base até valores muito altos, que geralmente correspondem a situações de infactibilidade / Abstract: This thesis presents a practical method for calculating power systems¿ maximum loading points (MLP) based on the geometric properties of the problem, an on the use of a Newton-Raphson load flow with step size optimization (LFSSO). The MLP is obtained by considering an acceptable error margin ?%, where this margin is a practical information that can be defined by the network operator. A formulation for practical application of the method to previously defined areas is also presented. Simulation results are presented for IEEE test transmission systems, distribution systems and realistic systems including the National Interconnected System (NIS) with over 4,500 buses. A large number of simulations have been carried out for evaluating its robustness, with acceptable errors ranging from 0.1% to 10%, and initial loadings starting from the base case up to very high values, which generally correspond to infeasibility situations / Mestrado / Energia Eletrica / Mestre em Engenharia Elétrica

Orientador: Rubén Augusto Romero Lázaro / Banca: José Roberto Sanches Mantovani / Banca: Sergio Azevedo de Oliveira / Banca: Eduardo Nobuhiro Asada / Banca: Carlos Roberto Mendonça da Rocha / Resumo: Neste trabalho, são propostas duas metodologias para resolver o problema de planejamento da expansão de sistemas de transmissão (PPEST). A primeira delas é uma apresentação de um algoritmo baseado em meta-heurística, Particle Swarm optimization (PSO), como ferramenta para resolver o PPEST estático. Nesta proposta são usados os modelos matemáticos de transporte e CC. Ambos os modelos foram implementados na linguagem de programação FORTRAN e o solver comercial MINOS foi usado para resolver os problemas de programação linear relacionados. A segunda proposta apresenta o PPEST considerando múltiplos cenários de geração usando o modelo linear disjuntivo. Este modelo foi implementado via AMPL e resolvido usando o solver comercial CPLEX. A fim de validar e qualificar a primeira metodologia proposta, cinco sistemas foram estudados, são eles: Garver, IEEE 24-barras, Sul Brasileiro, Colombiano e Norte-Nordeste Brasileiro. E para a segunda metodologia proposta, três sistemas foram estudados, são eles: Garver, IEEE 24-barras, Sul Brasileiro. Os resultados mostraram a eficiência e utilidade das duas metodologias propostas / Abstract: In this work, two methodologies to solve the transmission network expansion planning (TNEP) problem have been proposed. The first one is a presentation of a meta-heuristic based algorithm, Particle Swarm Optimization (PSO), as a tool to solve static TNEP problems. In this proposition, the mathematical models of transportation model and the DC model have been used. Both the models have been implemented via the programming language of FORTRAN and the commercial solver of MINOS has been used to solve the related LP problems. The second proposal, presents the TNEP problem considering multiple generation scenarios using the disjunctive linear model. This model has been implemented via AMPL and solved by the commercial solver of CPLEX. In order to validate and qualify the first proposed methodology, five case studies have been carried out, these cases are: Garver, IEEE 24-bus, Southern Brazil, Colombian, and Brazilian North-Northeast systems. And for the second proposed methodology, 3 case studies have been used such as: Garver, IEEE 24-bus and Southern Brazil test systems. The results show the effectiveness and the usefulness of the proposed methodologies / Doutor

Orientador: Rubén Augusto Romero Lázaro / Co-orientador: Marina lavorato de oliveira / Banca: José Roberto Sanches Mantovani / Banca: Carlos Roberto Mendonça da Rocha / Resumo: Neste trabalho, o algoritmo genético de Chu-Beasley especializado foi utilizado para resolver o problema de reconfiguração de sistemas de distribuição de energia elétrica, que é modelado como um problema de programação não linear inteiro misto. Um objetivo do trabalho é definir um espaço de busca reduzido, para reduzir o tempo computacional do algoritmo genético de Chu-Beasley especializado para resolver um problema de reconfiguração de sistemas de distribuição de energia elétrica. Para avaliar a factibilidade em relação às restrições de operação dos sistemas de energia elétrica foi implementado um algoritmo de fluxo de carga radial, para calcular o estado deste sistema e, a partir deste, as perdas de potência ativa e as correntes nos ramos. São apresentados testes realizados utilizando os sistemas de 14, 33, 84, 119 e 417 barras para avaliar a eficiência e robustez da metodologia proposta. Os resultados obtidos foram apresentados e comparados com os resultados encontrados na literatura / Abstract: In this paper, Chu-Beasley's specialized genetic algorithm will be used to solve the reconfiguration of distribuition system problem, which was modeled such as mixed integer nonlinear programming problem. In this work was defined a reduced search space to reduce the computational time of genetic algorithm of Chu-Beasley' specialized to solve the reconfiguration of distribution systems problem. To evaluate the feasibility regarding constraints of the electrical power systems' operation, an algorithm for radial load flow was implemented to calculate these nodal tensions of these systems and, losses and currents of the system. Tests are presented using the systems of 14, 33, 84, 119 and 417 buses to avaluate the efficiency and robustness of the proposed method. The results were presented and compared to those found in the literature / Mestre

Orientador: Rubén Augusto Romero Lázaro / Co-orientador: Marina Lavorato Oliveira / Banca: José Roberto Sanches Mantovani / Banca: Sergio Azevedo de Oliveira / Banca: Carlos Roberto Mendonça da Rocha / Banca: Luiz Gustavo Wesz da Silva / Resumo: A presente pesquisa apresenta uma análise teórica de alguns tópicos relevantes de programação linear relacionados ao método simplex. A motivação de analisar esses tópicos é torná-los mais didáticos e incentivar a aplicação dos mesmos na otimização de sistemas de energia elétrica e, em geral, na aplicação na otimização de problemas de engenharia. O método simplex em programação linear (PL) que foi por muito tempo a principal técnica de otimização de problemas de otimização lineares e não lineares (usado após a linearização) ainda é o mais utilizado na otimização de muitos problemas reais na engenharia e particularmente na engenharia elétrica. O método simplex resolve um problema de programação linear usando uma estratégia conceitualmente refinada. Para conhecer todas as versões do método simplex que podem ser usadas para resolver um problema de programação linear e também para dominar a parte conceitual do mesmo é necessário entender o ótimo de um problema de programação linear, quando um problema de programação linear é ilimitado, a lógica de otimização do método primal simplex, as condições nas quais um método simplex precisa de variáveis artificiais, o método primal simplex revisado, a teoria da dualidade em programação linear, a lógica de otimização do método dual simplex, a teoria da análise de sensibilidade e p ́os-otimizac ̧ão em programação linear, a lógica de otimização do método primal simplex canalizado bem como do método dual simplex canalizado. O produto desta pesquisa consiste em gerar um material didático que ajude de forma mais eficiente aos iniciantes em tópicos de pesquisa operacional. Desta forma, apresenta-se uma análise teórica e a reformulação de alguns tópicos relevantes relacionados com o método simplex na resolução de problemas de PL / Abstract: This research presents a theoretical analysis of some relevant topics related to linear programming via simplex method. The motivation of analyzing these topics makes them more didactic and easy to understand. As these kinds of methodologies are fast and unequivocal, they are applicable in various real-world engineering problems particularly in the field of power system optimization. In linear programming (LP), the simplex method has been the main technique to optimize the linear problem as well as the linearized problem (a problem with the nonlinear nature). The simplex method solves a linear programming problem using a conceptually refined strategy. In order to understand all of the available versions of the simplex method that can be used to find the solution of a linear programming problem and in order to have a detail study on them, it is necessary to understand: the optimality of such problems, where a linear programming problem is limited, the logic of optimization of the primal simplex method, in which condition a simplex method needs artificial variables, the revised primal simplex method, the duality theory in linear programming, the logic of optimization of the dual simplex methodology, the theory of sensitivity analysis and post-optimization in linear programming, and the logic of the primal or dual simplex for the boundary variables. The output of this research is to prepare a didactic reference and a user manual to help the beginner researchers in operations research. Therefore, a theoretical analysis and reformulation of some relevant topics related to the simplex method for solving LP problems is presented / Doutor