| Abstract (english) | For the Majority of the disturbances in the power system, responsible are nonlinear loads owned by customers (consumers), although supplier of electricity (grid operator - utility) can also contribute to these disturbances. The unwritten rule is that responsibility for maintaining the quality of voltage is on supplier of electricity, while customer (load) is responsible for the quality of the current. Supply voltage harmonics are mainly caused by higher nonlinear load current harmonics, which are connected to different voltage levels of the network. If there is a customer (load) with a negative effect on the network, producing significant amounts of higher harmonic currents, as result there is appearance of higher harmonics of the supply voltage at the point where a network is connected to the customer (point of common coupling). Disturbances in the form of reduced supply voltage quality, due to a significant share of higher voltage harmonics, reflect to all customers connected to the same connection point. Higher harmonics currents produced by nonlinear customer loads will be injected into all other appliances and customers connected to the same connection point. In this case, some customers may receive electrical energy with reduced quality. Since a supplier of electricity is responsible for the delivery of electricity within required quality, the supplier must limit the negative effect of individual customer. To limit the negative effect, determination of responsibility for such disturbances between customer and utility needs to be measured and detected. Power Quality is defined by the norms and standards that define the reference technical parameters and tolerances of delivered electrical energy. EN 50160 is a European standard that defines and describes the essential features of the voltage at the point of common coupling, at low and medium voltage network, under normal operating conditions. In today's networks, full of nonlinear loads which generate higher current harmonics, it is important to determine who is the real source of harmonic disturbances. Even if one side has linear load, due to resonant impedance of its network, it can amplify individual current harmonic which are coming from other side. Still, there is no standard procedure or method with which this problem can be easily resolved. Single point measurement methods will be discussed in detail in the doctoral thesis, and will examine the accuracy, functionality and applicability of these methods and propose the possibility of implementing certain methods and their algorithms on various types of measuring devices that are built into the power system. Detailed algorithms for all proposed methods in the MATLAB programming language will be developed, to examine their accuracy and the results of each method in different network conditions. To make this possible, it is necessary to create a computer model of the network by which it is possible to generate the desired current and voltage signals and check the results of determining the direction of higher harmonics of each method with known and expected results. The programming language used in this paper to create the network model and algorithm of each method is MATLAB. The main advantage of the MATLAB programming language is the advanced possibility of numerical and mathematical programming and the possibility of simulating the power system via the Simulink programming interface. It is very useful to easily process and visualize the results of processing the measured data. The development of computer models for simulation and determination of the direction of harmonic propagation using single-point measurement methods is divided into three basic steps. • Computer model for calculation of each method separately • Basic network model with and without power transformer • Distribution station model with different types of consumers After the measurements performed on the test system, the correctness of the method will be checked on the actual power system, within the power system of the Faculty of Electrical Engineering and Computing. On the main power supply for the building C of the Faculty of Electrical Engineering (FER) and Computing there is an analyzer of electricity quality, measurement accuracy class A. The device has the ability to record the wave signal of current and voltage with a reading frequency of 1024 samples per period. The device is remotely connected to the FER communication network and access to the device and metering data is provided through the software interface of the electricity quality monitoring system of the same manufacturer as the device itself. For the purposes of this work, a large number of wave signals were recorded at different loads and on different days of the week. The active power-based method typically shows accurate results under laboratory conditions, while under complex real-world conditions it shows results that may be erroneous (as in all phases of measurements on FER building C). The conclusion of this paper is that this method has not been shown to be completely reliable in a real system. In addition, a major disadvantage of this method is the required high level of measured data for the calculation of the method. Due to the high dependence on the exact angle of each higher voltage and current harmonic, ie due to the need to know this data, this method is applicable to network analyzers and advanced measuring terminals, which can measure the angle of each higher harmonic. The same input data is required for the active power-based method as for the RLC method. In this case, the RLC method proved to be a better and more accurate solution with more detailed results. The "distorted" harmonic force method results in data without a clear indication of which side contributes more to the direction of propagation of the higher harmonics. In order to be able to draw a conclusion, a great deal of experiential knowledge of the harmonic parameters of the network itself is necessary. It is necessary to observe other points in the system, the values of the results of the method on them and on the basis of reference measurements, confirmed by other methods, to determine which limit of the D / Db ratio determines the direction of propagation of higher harmonics. Given that the calculation of the "distorted" harmonic power method requires almost the same level of data from measuring devices as for the IEEE 1459-2000 method, it is concluded that a more applicable and useful method based on the IEEE 1459-2000 standard. The IEEE 1459-2000 method showed robustness and "relative accuracy" in selecting the direction of propagation of higher harmonics in all tests on laboratory examples and the actual system. "Relative accuracy" is stated because in cases where higher harmonics are present on both sides, the method shows mutual responsibility and in some borderline cases has shown an incorrect result. The disadvantage of this method is the impossibility of determining the responsibility for the propagation of each individual higher harmonic, so it shows only the general responsibility of each party. Although this method does not have a realistic application in determining the responsibility for individual complaints, the correct application of a number of measuring points of the method can make a significant contribution to determining the micro location of higher harmonic sources in the power system. The main advantage of this method is the low need for measured data, more precisely, the data required for the calculation are available on larger modern digital billing meters and on almost all basic measuring terminals with the possibility of remote communication. The RLC method showed accurate results in all measurements (according to the settings of the laboratory system) and it has the ability to determine in detail the influence of each side on the direction of propagation of each higher harmonic. Also, this method showed accuracy and high applicability on an actual system with a very complex situation (low THD and the presence of higher harmonic sources on both sides of the system). It is proposed to implement the RLC method at all technically applicable metering points in the power system and to apply the method when determining the responsibility of individual parties in complaints of users or system operators for negative effects of individual parties on the level of higher harmonics at the point of electricity collection. The proposal of this paper is the application of proposed methods within the smart grid system and within the network analyzers themselves (when possible). Some smart grids already have the option of harmonic analysis as part of the distribution network management system. Smart grids could collect a series of waveform recordings that can be requested at the request of the operator via the power quality monitoring system and advanced analyzers in the event of a THD increase above the recommended values. The waveform is tabulated with the number of rows equal to the number of reading frequencies in the period. The table needs to be read by the module for determining the direction of propagation of higher harmonics implemented in the smart grid system, which will use the algorithm to determine the direction of propagation of higher harmonics and display the desired data to the system operator. The presented system works in the same way using the MATLAB software tool used in this paper. For the purpose of processing the results from the actual system on FER's building C, the waveforms were collected via the power quality monitoring system and an advanced analyzer, transferred to an Excel spreadsheet read by the MATLAB software system and automatically applied to the method algorithm. The use of waveform and tabular input of waveform values allows to achieve the same accuracy of results regardless of the type and manufacturer of the network analyzer. Extensive laboratory testing of proposed methods in this thesis, show all pros and cons of each method, and applicability of each method to the real monitoring system. For further research, proposed methods should be tested on bigger system withing the power system grid, with testing done in various conditions through wider period. |
