Metod za efikasnu identifikaciju harmonijskih izobličenja u elektroenergetskoj mreži primenom modifikovanih elektronskih brojila

Abstract

The development of electronics has enabled the production of smart, energy-efficient appliances and equipment. Their application both in industry and in households has contributed to significant savings in electricity consumption and to CO2 emission reduction respectively. At first glance, it appears that significant progress has been made without any negative consequences. Unfortunately, that is not entirely true. Namely, modern electronic devices are based on direct current voltage (DC) and are supplied with alternating current (AC). The process of converting alternating current to direct current voltage is done by using the AC/DC converter. It is necessary to ensure minimum losses in the converter itself in order to increase the conversion efficiency. This is achieved when transistors operate in the switch mode. The current flowing through transistors is the strongest when transistor voltage is minimal and vice versa. Transistors operate at a frequency of several kHz. As a result of this operating mode, the device acts as an extremely nonlinear load which is manifested by pulse current energy consumption. From the standpoint of the utility system, this means that electricity does not follow the sinusoidal voltage waveform but that it is rich in harmonics. The harmonic components of the current flow through the utility and create additional harmonic voltage components at the final impedance lines. This leads to changes in the voltage waveform for all nearby consumers and the whole utility as well. In order to preserve the utility, the relevant international organizations have set standards which limit the maximum permissible value of the harmonics. The most famous standards in this field (IEEE 519-1992, IEC series 61000 and EN 50160) do not define the method which will be used to detect the dominant source of harmonic distortion. As a result, there are numerous methods in the current literature and practice that can be used for this purpose. Firstly, numerous adverse effects caused by the presence of harmonics in the electricity system were analyzed in this dissertation. Then, an overview of the existing methods for the identification of harmonic distortion sources was given. A thorough analysis on real examples has shown that the most widely used method based on the measurement of the active power harmonic components is not reliable. A common drawback of all the existing decisions is represented by a complicated implementation that disqualifies their application at the level of individual consumers (power meters). Therefore, it is mainly applied at the level of substations. A completely new method based on the measurement of power distortion is proposed in this dissertation in order to identify sources of harmonic distortion in the power network. The main advantage compared to competing solutions is the possibility to use it within the standard electronic meter. Moreover, the method can be implemented by modifying the standard meter at the software or hardware level. Irrespective of the type of implementation of this method, its application allows the distributor to receive information about the harmonic distortion source location directly at the level of consumers who caused them. The proposed method has been verified by simulations at a functional level and by measurements with the use of standard electronic meter and a personal computer. At the same time, the role of the personal computer can be transferred to the DSP, i.e. to the microprocessor part of the latest generation of meters through the software level intervention. Bearing in mind the fact that a large amount of meters of the old generation is used, the prototype of hardware device has been developed in order to extend the functions of the existing meters. The prototype was realised on the Cyclone 2 FPGA by the use of Altera DE2 development board. It enables the electricity distributor to identify the nonlinear load at the place of the connection to the network without replacing the existing electronic meters. A new generation of ASIC integrated Power Meters is currently being developed at the Faculty of Electronic Engineering LEDA Laboratory, University of Niš. The proposed method will be incorporated within the DSP block and realized in AMS 018mm CMOS technology.