| Sažetak (engleski) | The most significant characteristics of electric power system are operation reliability and safety. In order to ensure this, it is important to control the amount of reactive power in the system. Lack of required amount of reactive power in electric power system causes voltage drops which in some cases may cause a blackout of a part or even the entire electric power system, while the excessive amount of reactive power in electric power system causes voltage increase. In the case of excessive amount of reactive power, voltage can exceed maximum allowed values which affects the increased ageing of high voltage equipment. Main reactive power consumers in electric power system are loads which contain magnetic circuits such as asynchronous motors and transformers. Reactive power is needed for operation of such loads. Main producers of reactive power are synchronous generators, which govern the production of reactive power by their excitation system. High voltage overhead lines and underground cables also contribute to total reactive power production and consumption, depending on their operating condition. If transmission lines and cables are lightly loaded, then their capacitive character is dominating, and as a result they generate reactive power. The overproduction of reactive power is accompanied by an increase in voltages. On the other hand, in case when transmission lines and cables are highly loaded, their inductive character prevails. In this situation voltages drop, which can be compensated by reactive power sources. Issue of increased voltage and its regulation is briefly elaborated in chapter one of this thesis. Chapter one also describes the main producers (synchronous generators) and consumers (synchronous motors and transformers) of reactive power in system. In order to limit the voltage decrease caused by reactive power generation of the unloaded transmission lines, compensation devices are installed in certain network nodes (such as transformer stations and high voltage switchyards) which generate inductive reactive power and compensate capacitive reactive power of unloaded transmission lines. In this way, it is possible to maintain the voltage within the permissible boundaries. The second chapter describes the methods for keeping the voltage in high voltage transmission network within assigned boundaries, and possibilities of applying modern algorithms and methods such as genetic algorithms and neural networks for controlling voltage in electric power network. Neural networks have learning and independent decision possibilities, which allows them to be used for planning and operation managing of the electric power system. Therefore, they can also be used for solving the problems such as dynamic and static electric power system stability, alarm analysis and system surveillance, recognition of failure type and location, voltage and reactive power regulation, active power and frequency regulation. Advantages such as high speed of large input data analysis and short decision making time and disadvantages such as adaptation of input data to be suitable for calculations are described in this chapter. Different types of devices for reactive power compensation are described in the third chapter. A brief overview of reactive power compensation possibilities in electric power system is given, including the application of different dynamic compensation devices which control output reactive power at the point of their connection (such as synchronous generators, FACTS devices and synchronous compensators) and static compensation devices (such as high voltage shunt reactors, high voltage capacitors and regulation transformers). Fixed and variable shunt reactors (VSR) are briefly described in this chapter since the subject of this thesis is related to solving the issues of increased voltage by using high voltage variable shunt reactors and calculation of shunt reactor switching transients. Besides construction differences, the advantages of variable shunt reactors above fixed ones are described such as more precise voltage and reactive power regulation and lower electromechanical stresses on circuit breakers during energization and de-energization. The fourth chapter describes models of power system elements for calculation of power flows and electromagnetic transients. Power flow model consists of the following elements: generator, transformer, transmission lines, variable shunt reactors and loads. This model represents test network representing the electric power system with the following voltage levels: 400 kV, 220 kV and 110 kV. The power system elements were modelled at low frequencies (50 Hz) for power flows calculations and at high frequency range for calculation of switching overvoltages. Model for calculation of transients includes the model of high voltage switchyard with associated shunt reactor bay. Shunt reactor bay was modelled in detail taking into account the model of variable shunt reactor, disconnector, connecting conductors and circuit breaker (simplified and detailed model). Energization and de-energization was simulated in cases of minimal and maximal reactive power of variable shunt reactor. Simplified model of circuit breaker includes all electrical circuit breaker parameters except the electric arc while detailed circuit breaker model includes the model of electric arc. Electric arc model was represented with a black box model which describes the behaviour of the electric arc as a variable conductance depending on current and voltage in the electric network. Schwarz/Avdonin differential equation was used for modelling of the electric arc. For the calculation of inrush currents a simplified model of circuit breaker was used, while for de-energization both simplified and detailed circuit breaker models were used. SF6 circuit breaker with two breaking chambers for 400 kV network was modelled taking into account the interaction between the electric arc and transmission network. Based on conducted analysis and research, it could be concluded that a mentioned mathematical model for analysis of variable shunt reactor switching in high voltage transmission network represents a scientific contribution. The fifth chapter refers to method for selection of variable shunt reactor parameters (limitation method) which is based on power flow calculation and voltage sensitivity. The required minimum and maximum rated powers of the variable shunt reactor are determined by applying the proposed method. Minimum and maximum required powers of a variable shunt reactor are determined by applying the algorithm which determines the optimal power based on power flows and voltage sensitivity factor calculation taking into account the limitations of the network and equipment. It calculates needed regulating power of the shunt reactor to ensure that the voltage in all network nodes are within defined limitations. By applying previously described methods and algorithms the parameters of variable shunt reactor were selected in case of 400 kV test network at minimal and maximal system load. In order to take into account the real operating conditions of the transmission network, different operating scenarios were considered including minimum load condition (maximum reactive power of VSR is required) and maximum load condition (minimum reactive power is required). Based on calculation results a required range of reactive power regulation was determined which is sufficient for limiting voltages under maximum defined values. Based on provided research, developed method and conducted calculations in this chapter, the following scientific contribution was reached: method for selection of VSR parameters based on voltage sensitivity analysis in the transmission network. The sixth thesis chapter describes VSR switching transients in case of minimum and maximum reactive power. Inrush currents were calculated in case of VSR energization and overvoltages were calculated in case of de-energization (with and without reignition of electric arc inside the circuit breaker). Based on conducted simulations, it is shown that inrush currents of relatively large amplitudes can occur during VRS energization depending on the time instant of energization. De-energization of VSR is characterized by chopping of small inductive current before its neutral pass through the zero which is accompanied with the appearance of overvoltages. The mechanism of electric arc extinction in the circuit breaker chambers and development of transient recovery voltage between the circuit breaker contacts is a complex phenomenon. Overvoltages caused by VSR de-energization are the most severe in case when the amplitude of transient recovery voltage exceeds the dielectric strength between circuit breaker contacts after extinction of the electric arc (restrike occurrence). Dielectric stresses of the VSR insulation caused by restrike are similar to the stresses caused by standard lightning overvoltage waveform 1.2/50 μs. The calculations results showed that both inrush currents and switching overvoltages can be reduced significantly if a controlled switching is applied. Controlled switching requires special type of circuit breaker with a fast operating mechanism equipped with controlled switching device. In the seventh chapter an algorithm for optimal switching sequence of VSR was developed which reduces mechanical and electrical stresses of VSR and other equipment inside the substation. The proposed algorithm was applied on the test network and calculation results showed that minimum inrush currents can be achieved if the VSR is energized at maximum voltage value in case of VSR minimum reactive power (by applying the controlled switching). The minimum overvoltages appear in case when VSR is de-energized at maximum reactive power. Reignition of the electric arc can be prevented by using controlled switching. Installation of grading capacitors in parallel to breaking chambers of the circuit breaker contributes to equalization of the potential distribution across the chambers. Installation of surge arresters in the shunt reactor bay reduces the overvoltages on VSR and on other substation equipment. The most significant advantage of the this optimal switching sequence algorithm is its application during the planning of the transmission network development, when it is necessary to consider many possible operating scenarios in order to make the correct selection of the shunt reactor bay equipment (surge arresters and circuit breakers with the possibility of a controlled switching) and switching actions (switching time instant and corresponding reactive power) necessary to minimize the mechanical and electrical stresses of VSR and other equipment inside the substation. It can be concluded that the optimal switching sequence algorithm of high-voltage VSR with aim of reducing mechanical and electrical stress of high voltage equipment represents a scientific contribution. The increased voltage issue can be solved by installation of compensating devices (in this thesis it is VSR) in certain nodes of a test network. To make compensation more efficient, a method for selection of VSR parameters based on the sensitivity analysis of voltage in transmission network was developed and validated in the test network. As the reactive power changes in the network during the day, sometimes is necessary to switch on/off VSR which causes the inrush currents and switching overvoltages. A mathematical model of variable shunt reactor in high voltage transmission network was developed and applied for calculation of inrush currents and switching overvoltages. In order to achieve optimal switching sequence, an algorithm for preventing mechanical and electrical stress of VSR was developed which determines the range of possible inrush currents during VSR energization and overvoltages during its de-energization. Application of the proposed algorithm results with reducing of mechanical and electrical stresses of the equipment, preventing unwanted protection tripping, prolongation of equipment life time and increased equipment availability. |
